Search is not available for this dataset
id
string | submitter
string | authors
string | title
string | comments
string | journal-ref
string | doi
string | report-no
string | categories
string | license
string | abstract
string | versions
list | update_date
timestamp[s] | authors_parsed
sequence | prompt
string |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2504.04277 | Marios Kokkodis | Marios Kokkodis, Richard Demsyn-Jones, and Vijay Raghavan | Beyond the Hype: Embeddings vs. Prompting for Multiclass Classification
Tasks | null | null | null | null | cs.LG cs.AI cs.CL stat.AP | http://creativecommons.org/licenses/by/4.0/ | Are traditional classification approaches irrelevant in this era of AI hype?
We show that there are multiclass classification problems where predictive
models holistically outperform LLM prompt-based frameworks. Given text and
images from home-service project descriptions provided by Thumbtack customers,
we build embeddings-based softmax models that predict the professional category
(e.g., handyman, bathroom remodeling) associated with each problem description.
We then compare against prompts that ask state-of-the-art LLM models to solve
the same problem. We find that the embeddings approach outperforms the best LLM
prompts in terms of accuracy, calibration, latency, and financial cost. In
particular, the embeddings approach has 49.5% higher accuracy than the
prompting approach, and its superiority is consistent across text-only,
image-only, and text-image problem descriptions. Furthermore, it yields
well-calibrated probabilities, which we later use as confidence signals to
provide contextualized user experience during deployment. On the contrary,
prompting scores are overly uninformative. Finally, the embeddings approach is
14 and 81 times faster than prompting in processing images and text
respectively, while under realistic deployment assumptions, it can be up to 10
times cheaper. Based on these results, we deployed a variation of the
embeddings approach, and through A/B testing we observed performance consistent
with our offline analysis. Our study shows that for multiclass classification
problems that can leverage proprietary datasets, an embeddings-based approach
may yield unequivocally better results. Hence, scientists, practitioners,
engineers, and business leaders can use our study to go beyond the hype and
consider appropriate predictive models for their classification use cases.
| [
{
"version": "v1",
"created": "Sat, 5 Apr 2025 20:35:54 GMT"
},
{
"version": "v2",
"created": "Wed, 9 Apr 2025 17:15:47 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Kokkodis",
"Marios",
""
],
[
"Demsyn-Jones",
"Richard",
""
],
[
"Raghavan",
"Vijay",
""
]
] | TITLE: Beyond the Hype: Embeddings vs. Prompting for Multiclass Classification
Tasks
ABSTRACT: Are traditional classification approaches irrelevant in this era of AI hype?
We show that there are multiclass classification problems where predictive
models holistically outperform LLM prompt-based frameworks. Given text and
images from home-service project descriptions provided by Thumbtack customers,
we build embeddings-based softmax models that predict the professional category
(e.g., handyman, bathroom remodeling) associated with each problem description.
We then compare against prompts that ask state-of-the-art LLM models to solve
the same problem. We find that the embeddings approach outperforms the best LLM
prompts in terms of accuracy, calibration, latency, and financial cost. In
particular, the embeddings approach has 49.5% higher accuracy than the
prompting approach, and its superiority is consistent across text-only,
image-only, and text-image problem descriptions. Furthermore, it yields
well-calibrated probabilities, which we later use as confidence signals to
provide contextualized user experience during deployment. On the contrary,
prompting scores are overly uninformative. Finally, the embeddings approach is
14 and 81 times faster than prompting in processing images and text
respectively, while under realistic deployment assumptions, it can be up to 10
times cheaper. Based on these results, we deployed a variation of the
embeddings approach, and through A/B testing we observed performance consistent
with our offline analysis. Our study shows that for multiclass classification
problems that can leverage proprietary datasets, an embeddings-based approach
may yield unequivocally better results. Hence, scientists, practitioners,
engineers, and business leaders can use our study to go beyond the hype and
consider appropriate predictive models for their classification use cases.
|
2504.04514 | Yao Tao | Yao Tao, Yehui Tang, Yun Wang, Mingjian Zhu, Hailin Hu, Yunhe Wang | Saliency-driven Dynamic Token Pruning for Large Language Models | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Despite the recent success of large language models (LLMs), LLMs are
particularly challenging in long-sequence inference scenarios due to the
quadratic computational complexity of the attention mechanism. Inspired by the
interpretability theory of feature attribution in neural network models, we
observe that not all tokens have the same contribution. Based on this
observation, we propose a novel token pruning framework, namely Saliency-driven
Dynamic Token Pruning (SDTP), to gradually and dynamically prune redundant
tokens based on the input context. Specifically, a lightweight saliency-driven
prediction module is designed to estimate the importance score of each token
with its hidden state, which is added to different layers of the LLM to
hierarchically prune redundant tokens. Furthermore, a ranking-based
optimization strategy is proposed to minimize the ranking divergence of the
saliency score and the predicted importance score. Extensive experiments have
shown that our framework is generalizable to various models and datasets. By
hierarchically pruning 65\% of the input tokens, our method greatly reduces
33\% $\sim$ 47\% FLOPs and achieves speedup up to 1.75$\times$ during
inference, while maintaining comparable performance. We further demonstrate
that SDTP can be combined with KV cache compression method for further
compression.
| [
{
"version": "v1",
"created": "Sun, 6 Apr 2025 15:15:07 GMT"
},
{
"version": "v2",
"created": "Wed, 9 Apr 2025 14:36:19 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Tao",
"Yao",
""
],
[
"Tang",
"Yehui",
""
],
[
"Wang",
"Yun",
""
],
[
"Zhu",
"Mingjian",
""
],
[
"Hu",
"Hailin",
""
],
[
"Wang",
"Yunhe",
""
]
] | TITLE: Saliency-driven Dynamic Token Pruning for Large Language Models
ABSTRACT: Despite the recent success of large language models (LLMs), LLMs are
particularly challenging in long-sequence inference scenarios due to the
quadratic computational complexity of the attention mechanism. Inspired by the
interpretability theory of feature attribution in neural network models, we
observe that not all tokens have the same contribution. Based on this
observation, we propose a novel token pruning framework, namely Saliency-driven
Dynamic Token Pruning (SDTP), to gradually and dynamically prune redundant
tokens based on the input context. Specifically, a lightweight saliency-driven
prediction module is designed to estimate the importance score of each token
with its hidden state, which is added to different layers of the LLM to
hierarchically prune redundant tokens. Furthermore, a ranking-based
optimization strategy is proposed to minimize the ranking divergence of the
saliency score and the predicted importance score. Extensive experiments have
shown that our framework is generalizable to various models and datasets. By
hierarchically pruning 65\% of the input tokens, our method greatly reduces
33\% $\sim$ 47\% FLOPs and achieves speedup up to 1.75$\times$ during
inference, while maintaining comparable performance. We further demonstrate
that SDTP can be combined with KV cache compression method for further
compression.
|
2504.04713 | Yifei Yu | Yifei Yu, Qian-Wen Zhang, Lingfeng Qiao, Di Yin, Fang Li, Jie Wang,
Zengxi Chen, Suncong Zheng, Xiaolong Liang, Xing Sun | Sequential-NIAH: A Needle-In-A-Haystack Benchmark for Extracting
Sequential Needles from Long Contexts | null | null | null | null | cs.CL cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Evaluating the ability of large language models (LLMs) to handle extended
contexts is critical, particularly for retrieving information relevant to
specific queries embedded within lengthy inputs. We introduce Sequential-NIAH,
a benchmark specifically designed to evaluate the capability of LLMs to extract
sequential information items (known as needles) from long contexts. The
benchmark comprises three types of needle generation pipelines: synthetic,
real, and open-domain QA. It includes contexts ranging from 8K to 128K tokens
in length, with a dataset of 14,000 samples (2,000 reserved for testing). To
facilitate evaluation on this benchmark, we trained a synthetic data-driven
evaluation model capable of evaluating answer correctness based on
chronological or logical order, achieving an accuracy of 99.49% on synthetic
test data. We conducted experiments on six well-known LLMs, revealing that even
the best-performing model achieved a maximum accuracy of only 63.15%. Further
analysis highlights the growing challenges posed by increasing context lengths
and the number of needles, underscoring substantial room for improvement.
Additionally, noise robustness experiments validate the reliability of the
benchmark, making Sequential-NIAH an important reference for advancing research
on long text extraction capabilities of LLMs.
| [
{
"version": "v1",
"created": "Mon, 7 Apr 2025 03:50:12 GMT"
},
{
"version": "v2",
"created": "Wed, 9 Apr 2025 08:15:21 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Yu",
"Yifei",
""
],
[
"Zhang",
"Qian-Wen",
""
],
[
"Qiao",
"Lingfeng",
""
],
[
"Yin",
"Di",
""
],
[
"Li",
"Fang",
""
],
[
"Wang",
"Jie",
""
],
[
"Chen",
"Zengxi",
""
],
[
"Zheng",
"Suncong",
""
],
[
"Liang",
"Xiaolong",
""
],
[
"Sun",
"Xing",
""
]
] | TITLE: Sequential-NIAH: A Needle-In-A-Haystack Benchmark for Extracting
Sequential Needles from Long Contexts
ABSTRACT: Evaluating the ability of large language models (LLMs) to handle extended
contexts is critical, particularly for retrieving information relevant to
specific queries embedded within lengthy inputs. We introduce Sequential-NIAH,
a benchmark specifically designed to evaluate the capability of LLMs to extract
sequential information items (known as needles) from long contexts. The
benchmark comprises three types of needle generation pipelines: synthetic,
real, and open-domain QA. It includes contexts ranging from 8K to 128K tokens
in length, with a dataset of 14,000 samples (2,000 reserved for testing). To
facilitate evaluation on this benchmark, we trained a synthetic data-driven
evaluation model capable of evaluating answer correctness based on
chronological or logical order, achieving an accuracy of 99.49% on synthetic
test data. We conducted experiments on six well-known LLMs, revealing that even
the best-performing model achieved a maximum accuracy of only 63.15%. Further
analysis highlights the growing challenges posed by increasing context lengths
and the number of needles, underscoring substantial room for improvement.
Additionally, noise robustness experiments validate the reliability of the
benchmark, making Sequential-NIAH an important reference for advancing research
on long text extraction capabilities of LLMs.
|
2504.04798 | Jacob Si | Jacob Si, Zijing Ou, Mike Qu, Zhengrui Xiang, Yingzhen Li | TabRep: a Simple and Effective Continuous Representation for Training
Tabular Diffusion Models | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Diffusion models have been the predominant generative model for tabular data
generation. However, they face the conundrum of modeling under a separate
versus a unified data representation. The former encounters the challenge of
jointly modeling all multi-modal distributions of tabular data in one model.
While the latter alleviates this by learning a single representation for all
features, it currently leverages sparse suboptimal encoding heuristics and
necessitates additional computation costs. In this work, we address the latter
by presenting TabRep, a tabular diffusion architecture trained with a unified
continuous representation. To motivate the design of our representation, we
provide geometric insights into how the data manifold affects diffusion models.
The key attributes of our representation are composed of its density,
flexibility to provide ample separability for nominal features, and ability to
preserve intrinsic relationships. Ultimately, TabRep provides a simple yet
effective approach for training tabular diffusion models under a continuous
data manifold. Our results showcase that TabRep achieves superior performance
across a broad suite of evaluations. It is the first to synthesize tabular data
that exceeds the downstream quality of the original datasets while preserving
privacy and remaining computationally efficient.
| [
{
"version": "v1",
"created": "Mon, 7 Apr 2025 07:44:27 GMT"
},
{
"version": "v2",
"created": "Tue, 8 Apr 2025 15:10:24 GMT"
},
{
"version": "v3",
"created": "Wed, 9 Apr 2025 15:38:00 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Si",
"Jacob",
""
],
[
"Ou",
"Zijing",
""
],
[
"Qu",
"Mike",
""
],
[
"Xiang",
"Zhengrui",
""
],
[
"Li",
"Yingzhen",
""
]
] | TITLE: TabRep: a Simple and Effective Continuous Representation for Training
Tabular Diffusion Models
ABSTRACT: Diffusion models have been the predominant generative model for tabular data
generation. However, they face the conundrum of modeling under a separate
versus a unified data representation. The former encounters the challenge of
jointly modeling all multi-modal distributions of tabular data in one model.
While the latter alleviates this by learning a single representation for all
features, it currently leverages sparse suboptimal encoding heuristics and
necessitates additional computation costs. In this work, we address the latter
by presenting TabRep, a tabular diffusion architecture trained with a unified
continuous representation. To motivate the design of our representation, we
provide geometric insights into how the data manifold affects diffusion models.
The key attributes of our representation are composed of its density,
flexibility to provide ample separability for nominal features, and ability to
preserve intrinsic relationships. Ultimately, TabRep provides a simple yet
effective approach for training tabular diffusion models under a continuous
data manifold. Our results showcase that TabRep achieves superior performance
across a broad suite of evaluations. It is the first to synthesize tabular data
that exceeds the downstream quality of the original datasets while preserving
privacy and remaining computationally efficient.
|
2504.05523 | Elisabeth Fittschen | Elisabeth Fittschen, Sabrina Li, Tom Lippincott, Leshem Choshen, Craig
Messner | Pretraining Language Models for Diachronic Linguistic Change Discovery | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Large language models (LLMs) have shown potential as tools for scientific
discovery. This has engendered growing interest in their use in humanistic
disciplines, such as historical linguistics and literary studies. These fields
often construct arguments on the basis of delineations like genre, or more
inflexibly, time period. Although efforts have been made to restrict inference
to specific domains via fine-tuning or model editing, we posit that the only
true guarantee is domain-restricted pretraining -- typically, a data- and
compute-expensive proposition.
We show that efficient pretraining techniques can produce useful models over
corpora too large for easy manual inspection but too small for "typical" LLM
approaches. We employ a novel date-attribution pipeline in order to obtain a
temporally-segmented dataset of five 10-million-word slices. We train two
corresponding five-model batteries over these corpus segments, efficient
pretraining and Llama3-8B parameter efficiently finetuned.
We find that the pretrained models are faster to train than the finetuned
baselines and that they better respect the historical divisions of our corpus.
Emphasizing speed and precision over a-historical comprehensiveness enables a
number of novel approaches to hypothesis discovery and testing in our target
fields. Taking up diachronic linguistics as a testbed, we show that our method
enables the detection of a diverse set of phenomena, including en masse lexical
change, non-lexical (grammatical and morphological) change, and word sense
introduction/obsolescence. We provide a ready-to-use pipeline that allows
extension of our approach to other target fields with only minimal adaptation.
| [
{
"version": "v1",
"created": "Mon, 7 Apr 2025 21:51:32 GMT"
},
{
"version": "v2",
"created": "Wed, 9 Apr 2025 13:09:06 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Fittschen",
"Elisabeth",
""
],
[
"Li",
"Sabrina",
""
],
[
"Lippincott",
"Tom",
""
],
[
"Choshen",
"Leshem",
""
],
[
"Messner",
"Craig",
""
]
] | TITLE: Pretraining Language Models for Diachronic Linguistic Change Discovery
ABSTRACT: Large language models (LLMs) have shown potential as tools for scientific
discovery. This has engendered growing interest in their use in humanistic
disciplines, such as historical linguistics and literary studies. These fields
often construct arguments on the basis of delineations like genre, or more
inflexibly, time period. Although efforts have been made to restrict inference
to specific domains via fine-tuning or model editing, we posit that the only
true guarantee is domain-restricted pretraining -- typically, a data- and
compute-expensive proposition.
We show that efficient pretraining techniques can produce useful models over
corpora too large for easy manual inspection but too small for "typical" LLM
approaches. We employ a novel date-attribution pipeline in order to obtain a
temporally-segmented dataset of five 10-million-word slices. We train two
corresponding five-model batteries over these corpus segments, efficient
pretraining and Llama3-8B parameter efficiently finetuned.
We find that the pretrained models are faster to train than the finetuned
baselines and that they better respect the historical divisions of our corpus.
Emphasizing speed and precision over a-historical comprehensiveness enables a
number of novel approaches to hypothesis discovery and testing in our target
fields. Taking up diachronic linguistics as a testbed, we show that our method
enables the detection of a diverse set of phenomena, including en masse lexical
change, non-lexical (grammatical and morphological) change, and word sense
introduction/obsolescence. We provide a ready-to-use pipeline that allows
extension of our approach to other target fields with only minimal adaptation.
|
2504.05643 | Kaiji Sekimoto | Kaiji Sekimoto and Muneki Yasuda | Effective Method for Inverse Ising Problem under Missing Observations in
Restricted Boltzmann Machines | null | null | null | null | stat.ML cond-mat.dis-nn cs.LG physics.data-an | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Restricted Boltzmann machines (RBMs) are energy-based models analogous to the
Ising model and are widely applied in statistical machine learning. The
standard inverse Ising problem with a complete dataset requires computing both
data and model expectations and is computationally challenging because model
expectations have a combinatorial explosion. Furthermore, in many applications,
the available datasets are partially incomplete, making it difficult to compute
even data expectations. In this study, we propose a approximation framework for
these expectations in the practical inverse Ising problems that integrates
mean-field approximation or persistent contrastive divergence to generate
refined initial points and spatial Monte Carlo integration to enhance estimator
accuracy. We demonstrate that the proposed method effectively and accurately
tunes the model parameters in comparison to the conventional method.
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 03:39:56 GMT"
},
{
"version": "v2",
"created": "Wed, 9 Apr 2025 06:05:02 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Sekimoto",
"Kaiji",
""
],
[
"Yasuda",
"Muneki",
""
]
] | TITLE: Effective Method for Inverse Ising Problem under Missing Observations in
Restricted Boltzmann Machines
ABSTRACT: Restricted Boltzmann machines (RBMs) are energy-based models analogous to the
Ising model and are widely applied in statistical machine learning. The
standard inverse Ising problem with a complete dataset requires computing both
data and model expectations and is computationally challenging because model
expectations have a combinatorial explosion. Furthermore, in many applications,
the available datasets are partially incomplete, making it difficult to compute
even data expectations. In this study, we propose a approximation framework for
these expectations in the practical inverse Ising problems that integrates
mean-field approximation or persistent contrastive divergence to generate
refined initial points and spatial Monte Carlo integration to enhance estimator
accuracy. We demonstrate that the proposed method effectively and accurately
tunes the model parameters in comparison to the conventional method.
|
2504.05759 | Nathana\"el Beau | Nathana\"el Beau and Beno\^it Crabb\'e | RETROcode: Leveraging a Code Database for Improved Natural Language to
Code Generation | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by-nc-nd/4.0/ | As text and code resources have expanded, large-scale pre-trained models have
shown promising capabilities in code generation tasks, typically employing
supervised fine-tuning with problem statement-program pairs. However,
increasing model size and data volume for performance gains also raises
computational demands and risks of overfitting. Addressing these challenges, we
present RETROcode, a novel adaptation of the RETRO architecture \cite{RETRO}
for sequence-to-sequence models, utilizing a large code database as an
auxiliary scaling method. This approach, diverging from simply enlarging model
and dataset sizes, allows RETROcode to leverage a vast code database for
prediction, enhancing the model's efficiency by integrating extensive memory.
Our findings indicate that RETROcode not only outperforms similar-sized
traditional architectures on test sets but also approaches the effectiveness of
the much larger Codex model, despite being trained from scratch on a
substantially smaller dataset.
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 07:41:13 GMT"
},
{
"version": "v2",
"created": "Wed, 9 Apr 2025 06:55:15 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Beau",
"Nathanaël",
""
],
[
"Crabbé",
"Benoît",
""
]
] | TITLE: RETROcode: Leveraging a Code Database for Improved Natural Language to
Code Generation
ABSTRACT: As text and code resources have expanded, large-scale pre-trained models have
shown promising capabilities in code generation tasks, typically employing
supervised fine-tuning with problem statement-program pairs. However,
increasing model size and data volume for performance gains also raises
computational demands and risks of overfitting. Addressing these challenges, we
present RETROcode, a novel adaptation of the RETRO architecture \cite{RETRO}
for sequence-to-sequence models, utilizing a large code database as an
auxiliary scaling method. This approach, diverging from simply enlarging model
and dataset sizes, allows RETROcode to leverage a vast code database for
prediction, enhancing the model's efficiency by integrating extensive memory.
Our findings indicate that RETROcode not only outperforms similar-sized
traditional architectures on test sets but also approaches the effectiveness of
the much larger Codex model, despite being trained from scratch on a
substantially smaller dataset.
|
2504.05795 | Yanping Zha | Hao Zhang, Yanping Zha, Qingwei Zhuang, Zhenfeng Shao, Jiayi Ma | Robust Fusion Controller: Degradation-aware Image Fusion with
Fine-grained Language Instructions | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Current image fusion methods struggle to adapt to real-world environments
encompassing diverse degradations with spatially varying characteristics. To
address this challenge, we propose a robust fusion controller (RFC) capable of
achieving degradation-aware image fusion through fine-grained language
instructions, ensuring its reliable application in adverse environments.
Specifically, RFC first parses language instructions to innovatively derive the
functional condition and the spatial condition, where the former specifies the
degradation type to remove, while the latter defines its spatial coverage.
Then, a composite control priori is generated through a multi-condition
coupling network, achieving a seamless transition from abstract language
instructions to latent control variables. Subsequently, we design a hybrid
attention-based fusion network to aggregate multi-modal information, in which
the obtained composite control priori is deeply embedded to linearly modulate
the intermediate fused features. To ensure the alignment between language
instructions and control outcomes, we introduce a novel language-feature
alignment loss, which constrains the consistency between feature-level gains
and the composite control priori. Extensive experiments on publicly available
datasets demonstrate that our RFC is robust against various composite
degradations, particularly in highly challenging flare scenarios.
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 08:22:55 GMT"
},
{
"version": "v2",
"created": "Wed, 9 Apr 2025 10:05:59 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Zhang",
"Hao",
""
],
[
"Zha",
"Yanping",
""
],
[
"Zhuang",
"Qingwei",
""
],
[
"Shao",
"Zhenfeng",
""
],
[
"Ma",
"Jiayi",
""
]
] | TITLE: Robust Fusion Controller: Degradation-aware Image Fusion with
Fine-grained Language Instructions
ABSTRACT: Current image fusion methods struggle to adapt to real-world environments
encompassing diverse degradations with spatially varying characteristics. To
address this challenge, we propose a robust fusion controller (RFC) capable of
achieving degradation-aware image fusion through fine-grained language
instructions, ensuring its reliable application in adverse environments.
Specifically, RFC first parses language instructions to innovatively derive the
functional condition and the spatial condition, where the former specifies the
degradation type to remove, while the latter defines its spatial coverage.
Then, a composite control priori is generated through a multi-condition
coupling network, achieving a seamless transition from abstract language
instructions to latent control variables. Subsequently, we design a hybrid
attention-based fusion network to aggregate multi-modal information, in which
the obtained composite control priori is deeply embedded to linearly modulate
the intermediate fused features. To ensure the alignment between language
instructions and control outcomes, we introduce a novel language-feature
alignment loss, which constrains the consistency between feature-level gains
and the composite control priori. Extensive experiments on publicly available
datasets demonstrate that our RFC is robust against various composite
degradations, particularly in highly challenging flare scenarios.
|
2504.06122 | Yahui Liu | Jingyuan Zhang, Qi Wang, Xingguang Ji, Yahui Liu, Yang Yue, Fuzheng
Zhang, Di Zhang, Guorui Zhou, Kun Gai | Leanabell-Prover: Posttraining Scaling in Formal Reasoning | 23 pages, 6 figures | null | null | null | cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent advances in automated theorem proving (ATP) through LLMs have
highlighted the potential of formal reasoning with Lean 4 codes. However, ATP
has not yet be revolutionized by the recent posttraining scaling as
demonstrated by Open AI O1/O3 and Deepseek R1. In this work, we investigate the
entire posttraining of ATP, aiming to align it with breakthroughs in reasoning
models in natural languages. To begin, we continual train current ATP models
with a hybrid dataset, which consists of numerous statement-proof pairs, and
additional data aimed at incorporating cognitive behaviors that emulate human
reasoning and hypothesis refinement. Next, we explore reinforcement learning
with the use of outcome reward returned by Lean 4 compiler. Through our
designed continual training and reinforcement learning processes, we have
successfully improved existing formal provers, including both
DeepSeek-Prover-v1.5 and Goedel-Prover, achieving state-of-the-art performance
in the field of whole-proof generation. For example, we achieve a 59.8% pass
rate (pass@32) on MiniF2F. This is an on-going project and we will
progressively update our findings, release our data and training details.
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 15:15:26 GMT"
},
{
"version": "v2",
"created": "Wed, 9 Apr 2025 04:03:00 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Zhang",
"Jingyuan",
""
],
[
"Wang",
"Qi",
""
],
[
"Ji",
"Xingguang",
""
],
[
"Liu",
"Yahui",
""
],
[
"Yue",
"Yang",
""
],
[
"Zhang",
"Fuzheng",
""
],
[
"Zhang",
"Di",
""
],
[
"Zhou",
"Guorui",
""
],
[
"Gai",
"Kun",
""
]
] | TITLE: Leanabell-Prover: Posttraining Scaling in Formal Reasoning
ABSTRACT: Recent advances in automated theorem proving (ATP) through LLMs have
highlighted the potential of formal reasoning with Lean 4 codes. However, ATP
has not yet be revolutionized by the recent posttraining scaling as
demonstrated by Open AI O1/O3 and Deepseek R1. In this work, we investigate the
entire posttraining of ATP, aiming to align it with breakthroughs in reasoning
models in natural languages. To begin, we continual train current ATP models
with a hybrid dataset, which consists of numerous statement-proof pairs, and
additional data aimed at incorporating cognitive behaviors that emulate human
reasoning and hypothesis refinement. Next, we explore reinforcement learning
with the use of outcome reward returned by Lean 4 compiler. Through our
designed continual training and reinforcement learning processes, we have
successfully improved existing formal provers, including both
DeepSeek-Prover-v1.5 and Goedel-Prover, achieving state-of-the-art performance
in the field of whole-proof generation. For example, we achieve a 59.8% pass
rate (pass@32) on MiniF2F. This is an on-going project and we will
progressively update our findings, release our data and training details.
|
2504.06125 | Luigi Tresca | Luigi Tresca, Carolin Schmidt, James Harrison, Filipe Rodrigues,
Gioele Zardini, Daniele Gammelli, and Marco Pavone | Robo-taxi Fleet Coordination at Scale via Reinforcement Learning | 12 pages, 6 figures, 6 tables | null | null | null | cs.LG cs.SY eess.SY | http://creativecommons.org/licenses/by/4.0/ | Fleets of robo-taxis offering on-demand transportation services, commonly
known as Autonomous Mobility-on-Demand (AMoD) systems, hold significant promise
for societal benefits, such as reducing pollution, energy consumption, and
urban congestion. However, orchestrating these systems at scale remains a
critical challenge, with existing coordination algorithms often failing to
exploit the systems' full potential. This work introduces a novel
decision-making framework that unites mathematical modeling with data-driven
techniques. In particular, we present the AMoD coordination problem through the
lens of reinforcement learning and propose a graph network-based framework that
exploits the main strengths of graph representation learning, reinforcement
learning, and classical operations research tools. Extensive evaluations across
diverse simulation fidelities and scenarios demonstrate the flexibility of our
approach, achieving superior system performance, computational efficiency, and
generalizability compared to prior methods. Finally, motivated by the need to
democratize research efforts in this area, we release publicly available
benchmarks, datasets, and simulators for network-level coordination alongside
an open-source codebase designed to provide accessible simulation platforms and
establish a standardized validation process for comparing methodologies. Code
available at: https://github.com/StanfordASL/RL4AMOD
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 15:19:41 GMT"
},
{
"version": "v2",
"created": "Wed, 9 Apr 2025 07:54:20 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Tresca",
"Luigi",
""
],
[
"Schmidt",
"Carolin",
""
],
[
"Harrison",
"James",
""
],
[
"Rodrigues",
"Filipe",
""
],
[
"Zardini",
"Gioele",
""
],
[
"Gammelli",
"Daniele",
""
],
[
"Pavone",
"Marco",
""
]
] | TITLE: Robo-taxi Fleet Coordination at Scale via Reinforcement Learning
ABSTRACT: Fleets of robo-taxis offering on-demand transportation services, commonly
known as Autonomous Mobility-on-Demand (AMoD) systems, hold significant promise
for societal benefits, such as reducing pollution, energy consumption, and
urban congestion. However, orchestrating these systems at scale remains a
critical challenge, with existing coordination algorithms often failing to
exploit the systems' full potential. This work introduces a novel
decision-making framework that unites mathematical modeling with data-driven
techniques. In particular, we present the AMoD coordination problem through the
lens of reinforcement learning and propose a graph network-based framework that
exploits the main strengths of graph representation learning, reinforcement
learning, and classical operations research tools. Extensive evaluations across
diverse simulation fidelities and scenarios demonstrate the flexibility of our
approach, achieving superior system performance, computational efficiency, and
generalizability compared to prior methods. Finally, motivated by the need to
democratize research efforts in this area, we release publicly available
benchmarks, datasets, and simulators for network-level coordination alongside
an open-source codebase designed to provide accessible simulation platforms and
establish a standardized validation process for comparing methodologies. Code
available at: https://github.com/StanfordASL/RL4AMOD
|
2504.06160 | Rijul Magu | Rijul Magu, Arka Dutta, Sean Kim, Ashiqur R. KhudaBukhsh, Munmun De
Choudhury | Navigating the Rabbit Hole: Emergent Biases in LLM-Generated Attack
Narratives Targeting Mental Health Groups | null | null | null | null | cs.CL cs.AI cs.CY cs.LG cs.SI | http://creativecommons.org/licenses/by/4.0/ | Large Language Models (LLMs) have been shown to demonstrate imbalanced biases
against certain groups. However, the study of unprovoked targeted attacks by
LLMs towards at-risk populations remains underexplored. Our paper presents
three novel contributions: (1) the explicit evaluation of LLM-generated attacks
on highly vulnerable mental health groups; (2) a network-based framework to
study the propagation of relative biases; and (3) an assessment of the relative
degree of stigmatization that emerges from these attacks. Our analysis of a
recently released large-scale bias audit dataset reveals that mental health
entities occupy central positions within attack narrative networks, as revealed
by a significantly higher mean centrality of closeness (p-value = 4.06e-10) and
dense clustering (Gini coefficient = 0.7). Drawing from sociological
foundations of stigmatization theory, our stigmatization analysis indicates
increased labeling components for mental health disorder-related targets
relative to initial targets in generation chains. Taken together, these
insights shed light on the structural predilections of large language models to
heighten harmful discourse and highlight the need for suitable approaches for
mitigation.
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 15:56:57 GMT"
},
{
"version": "v2",
"created": "Wed, 9 Apr 2025 04:24:38 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Magu",
"Rijul",
""
],
[
"Dutta",
"Arka",
""
],
[
"Kim",
"Sean",
""
],
[
"KhudaBukhsh",
"Ashiqur R.",
""
],
[
"De Choudhury",
"Munmun",
""
]
] | TITLE: Navigating the Rabbit Hole: Emergent Biases in LLM-Generated Attack
Narratives Targeting Mental Health Groups
ABSTRACT: Large Language Models (LLMs) have been shown to demonstrate imbalanced biases
against certain groups. However, the study of unprovoked targeted attacks by
LLMs towards at-risk populations remains underexplored. Our paper presents
three novel contributions: (1) the explicit evaluation of LLM-generated attacks
on highly vulnerable mental health groups; (2) a network-based framework to
study the propagation of relative biases; and (3) an assessment of the relative
degree of stigmatization that emerges from these attacks. Our analysis of a
recently released large-scale bias audit dataset reveals that mental health
entities occupy central positions within attack narrative networks, as revealed
by a significantly higher mean centrality of closeness (p-value = 4.06e-10) and
dense clustering (Gini coefficient = 0.7). Drawing from sociological
foundations of stigmatization theory, our stigmatization analysis indicates
increased labeling components for mental health disorder-related targets
relative to initial targets in generation chains. Taken together, these
insights shed light on the structural predilections of large language models to
heighten harmful discourse and highlight the need for suitable approaches for
mitigation.
|
2504.06270 | Wenqiao Zhu | Wenqiao Zhu, Lulu Wang, Jun Wu | Addressing Cold-start Problem in Click-Through Rate Prediction via
Supervised Diffusion Modeling | null | null | null | null | cs.IR cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Predicting Click-Through Rates is a crucial function within recommendation
and advertising platforms, as the output of CTR prediction determines the order
of items shown to users. The Embedding \& MLP paradigm has become a standard
approach for industrial recommendation systems and has been widely deployed.
However, this paradigm suffers from cold-start problems, where there is either
no or only limited user action data available, leading to poorly learned ID
embeddings. The cold-start problem hampers the performance of new items. To
address this problem, we designed a novel diffusion model to generate a
warmed-up embedding for new items. Specifically, we define a novel diffusion
process between the ID embedding space and the side information space. In
addition, we can derive a sub-sequence from the diffusion steps to expedite
training, given that our diffusion model is non-Markovian. Our diffusion model
is supervised by both the variational inference and binary cross-entropy
objectives, enabling it to generate warmed-up embeddings for items in both the
cold-start and warm-up phases. Additionally, we have conducted extensive
experiments on three recommendation datasets. The results confirmed the
effectiveness of our approach.
| [
{
"version": "v1",
"created": "Sat, 1 Mar 2025 13:43:06 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Zhu",
"Wenqiao",
""
],
[
"Wang",
"Lulu",
""
],
[
"Wu",
"Jun",
""
]
] | TITLE: Addressing Cold-start Problem in Click-Through Rate Prediction via
Supervised Diffusion Modeling
ABSTRACT: Predicting Click-Through Rates is a crucial function within recommendation
and advertising platforms, as the output of CTR prediction determines the order
of items shown to users. The Embedding \& MLP paradigm has become a standard
approach for industrial recommendation systems and has been widely deployed.
However, this paradigm suffers from cold-start problems, where there is either
no or only limited user action data available, leading to poorly learned ID
embeddings. The cold-start problem hampers the performance of new items. To
address this problem, we designed a novel diffusion model to generate a
warmed-up embedding for new items. Specifically, we define a novel diffusion
process between the ID embedding space and the side information space. In
addition, we can derive a sub-sequence from the diffusion steps to expedite
training, given that our diffusion model is non-Markovian. Our diffusion model
is supervised by both the variational inference and binary cross-entropy
objectives, enabling it to generate warmed-up embeddings for items in both the
cold-start and warm-up phases. Additionally, we have conducted extensive
experiments on three recommendation datasets. The results confirmed the
effectiveness of our approach.
|
2504.06272 | Kevin Dela Rosa | Kevin Dela Rosa | RAVEN: An Agentic Framework for Multimodal Entity Discovery from
Large-Scale Video Collections | Presented at AI Agent for Information Retrieval: Generating and
Ranking (Agent4IR) @ AAAI 2025
[https://sites.google.com/view/ai4ir/aaai-2025] | null | null | null | cs.IR cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present RAVEN an adaptive AI agent framework designed for multimodal
entity discovery and retrieval in large-scale video collections. Synthesizing
information across visual, audio, and textual modalities, RAVEN autonomously
processes video data to produce structured, actionable representations for
downstream tasks. Key contributions include (1) a category understanding step
to infer video themes and general-purpose entities, (2) a schema generation
mechanism that dynamically defines domain-specific entities and attributes, and
(3) a rich entity extraction process that leverages semantic retrieval and
schema-guided prompting. RAVEN is designed to be model-agnostic, allowing the
integration of different vision-language models (VLMs) and large language
models (LLMs) based on application-specific requirements. This flexibility
supports diverse applications in personalized search, content discovery, and
scalable information retrieval, enabling practical applications across vast
datasets.
| [
{
"version": "v1",
"created": "Mon, 3 Mar 2025 08:28:58 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Rosa",
"Kevin Dela",
""
]
] | TITLE: RAVEN: An Agentic Framework for Multimodal Entity Discovery from
Large-Scale Video Collections
ABSTRACT: We present RAVEN an adaptive AI agent framework designed for multimodal
entity discovery and retrieval in large-scale video collections. Synthesizing
information across visual, audio, and textual modalities, RAVEN autonomously
processes video data to produce structured, actionable representations for
downstream tasks. Key contributions include (1) a category understanding step
to infer video themes and general-purpose entities, (2) a schema generation
mechanism that dynamically defines domain-specific entities and attributes, and
(3) a rich entity extraction process that leverages semantic retrieval and
schema-guided prompting. RAVEN is designed to be model-agnostic, allowing the
integration of different vision-language models (VLMs) and large language
models (LLMs) based on application-specific requirements. This flexibility
supports diverse applications in personalized search, content discovery, and
scalable information retrieval, enabling practical applications across vast
datasets.
|
2504.06274 | Ngoc Luyen Le | Ngoc Luyen Le, Marie-H\'el\`ene Abel | Joint Group Profiling and Recommendation via Deep Neural Network-based
Multi-Task Learning | null | null | null | null | cs.IR cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Group recommender systems aim to generate recommendations that align with the
collective preferences of a group, introducing challenges that differ
significantly from those in individual recommendation scenarios. This paper
presents Joint Group Profiling and Recommendation via Deep Neural Network-based
Multi-Task Learning, a framework that unifies group profiling and
recommendation tasks within a single model. By jointly learning these tasks,
the model develops a deeper understanding of group dynamics, leading to
improved recommendation accuracy. The shared representations between the two
tasks facilitate the discovery of latent features essential to both, resulting
in richer and more informative group embeddings. To further enhance
performance, an attention mechanism is integrated to dynamically evaluate the
relevance of different group features and item attributes, ensuring the model
prioritizes the most impactful information. Experiments and evaluations on
real-world datasets demonstrate that our multi-task learning approach
consistently outperforms baseline models in terms of accuracy, validating its
effectiveness and robustness.
| [
{
"version": "v1",
"created": "Wed, 5 Mar 2025 14:28:48 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Le",
"Ngoc Luyen",
""
],
[
"Abel",
"Marie-Hélène",
""
]
] | TITLE: Joint Group Profiling and Recommendation via Deep Neural Network-based
Multi-Task Learning
ABSTRACT: Group recommender systems aim to generate recommendations that align with the
collective preferences of a group, introducing challenges that differ
significantly from those in individual recommendation scenarios. This paper
presents Joint Group Profiling and Recommendation via Deep Neural Network-based
Multi-Task Learning, a framework that unifies group profiling and
recommendation tasks within a single model. By jointly learning these tasks,
the model develops a deeper understanding of group dynamics, leading to
improved recommendation accuracy. The shared representations between the two
tasks facilitate the discovery of latent features essential to both, resulting
in richer and more informative group embeddings. To further enhance
performance, an attention mechanism is integrated to dynamically evaluate the
relevance of different group features and item attributes, ensuring the model
prioritizes the most impactful information. Experiments and evaluations on
real-world datasets demonstrate that our multi-task learning approach
consistently outperforms baseline models in terms of accuracy, validating its
effectiveness and robustness.
|
2504.06282 | Jakub Vasicek | Jakub Va\v{s}\'i\v{c}ek, Dafni Skiadopoulou, Ksenia G. Kuznetsova,
Lukas K\"all, Marc Vaudel, Stefan Bruckner | ProHap Explorer: Visualizing Haplotypes in Proteogenomic Datasets | null | null | null | null | q-bio.GN cs.GR | http://creativecommons.org/licenses/by/4.0/ | In mass spectrometry-based proteomics, experts usually project data onto a
single set of reference sequences, overlooking the influence of common
haplotypes (combinations of genetic variants inherited together from a parent).
We recently introduced ProHap, a tool for generating customized protein
haplotype databases. Here, we present ProHap Explorer, a visualization
interface designed to investigate the influence of common haplotypes on the
human proteome. It enables users to explore haplotypes, their effects on
protein sequences, and the identification of non-canonical peptides in public
mass spectrometry datasets. The design builds on well-established
representations in biological sequence analysis, ensuring familiarity for
domain experts while integrating novel interactive elements tailored to
proteogenomic data exploration. User interviews with proteomics experts
confirmed the tool's utility, highlighting its ability to reveal whether
haplotypes affect proteins of interest. By facilitating the intuitive
exploration of proteogenomic variation, ProHap Explorer supports research in
personalized medicine and the development of targeted therapies.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 14:48:20 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Vašíček",
"Jakub",
""
],
[
"Skiadopoulou",
"Dafni",
""
],
[
"Kuznetsova",
"Ksenia G.",
""
],
[
"Käll",
"Lukas",
""
],
[
"Vaudel",
"Marc",
""
],
[
"Bruckner",
"Stefan",
""
]
] | TITLE: ProHap Explorer: Visualizing Haplotypes in Proteogenomic Datasets
ABSTRACT: In mass spectrometry-based proteomics, experts usually project data onto a
single set of reference sequences, overlooking the influence of common
haplotypes (combinations of genetic variants inherited together from a parent).
We recently introduced ProHap, a tool for generating customized protein
haplotype databases. Here, we present ProHap Explorer, a visualization
interface designed to investigate the influence of common haplotypes on the
human proteome. It enables users to explore haplotypes, their effects on
protein sequences, and the identification of non-canonical peptides in public
mass spectrometry datasets. The design builds on well-established
representations in biological sequence analysis, ensuring familiarity for
domain experts while integrating novel interactive elements tailored to
proteogenomic data exploration. User interviews with proteomics experts
confirmed the tool's utility, highlighting its ability to reveal whether
haplotypes affect proteins of interest. By facilitating the intuitive
exploration of proteogenomic variation, ProHap Explorer supports research in
personalized medicine and the development of targeted therapies.
|
2504.06285 | Bryar Hassan Dr. | Bryar A. Hassan, Shko M. Qader, Alla A. Hassan, Joan Lu, Aram M.
Ahmed, Jafar Majidpour, Tarik A. Rashid | Reducing Formal Context Extraction: A Newly Proposed Framework from Big
Corpora | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Automating the extraction of concept hierarchies from free text is
advantageous because manual generation is frequently labor- and
resource-intensive. Free result, the whole procedure for concept hierarchy
learning from free text entails several phases, including sentence-level text
processing, sentence splitting, and tokenization. Lemmatization is after formal
context analysis (FCA) to derive the pairings. Nevertheless, there could be a
few uninteresting and incorrect pairings in the formal context. It may take a
while to generate formal context; thus, size reduction formal context is
necessary to weed out irrelevant and incorrect pairings to extract the concept
lattice and hierarchies more quickly. This study aims to propose a framework
for reducing formal context in extracting concept hierarchies from free text to
reduce the ambiguity of the formal context. We achieve this by reducing the
size of the formal context using a hybrid of a WordNet-based method and a
frequency-based technique. Using 385 samples from the Wikipedia corpus and the
suggested framework, tests are carried out to examine the reduced size of
formal context, leading to concept lattice and concept hierarchy. With the help
of concept lattice-invariants, the generated formal context lattice is compared
to the normal one. In contrast to basic ones, the homomorphic between the
resultant lattices retains up to 98% of the quality of the generating concept
hierarchies, and the reduced concept lattice receives the structural connection
of the standard one. Additionally, the new framework is compared to five
baseline techniques to calculate the running time on random datasets with
various densities. The findings demonstrate that, in various fill ratios,
hybrid approaches of the proposed method outperform other indicated competing
strategies in concept lattice performance.
| [
{
"version": "v1",
"created": "Tue, 1 Apr 2025 09:24:07 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Hassan",
"Bryar A.",
""
],
[
"Qader",
"Shko M.",
""
],
[
"Hassan",
"Alla A.",
""
],
[
"Lu",
"Joan",
""
],
[
"Ahmed",
"Aram M.",
""
],
[
"Majidpour",
"Jafar",
""
],
[
"Rashid",
"Tarik A.",
""
]
] | TITLE: Reducing Formal Context Extraction: A Newly Proposed Framework from Big
Corpora
ABSTRACT: Automating the extraction of concept hierarchies from free text is
advantageous because manual generation is frequently labor- and
resource-intensive. Free result, the whole procedure for concept hierarchy
learning from free text entails several phases, including sentence-level text
processing, sentence splitting, and tokenization. Lemmatization is after formal
context analysis (FCA) to derive the pairings. Nevertheless, there could be a
few uninteresting and incorrect pairings in the formal context. It may take a
while to generate formal context; thus, size reduction formal context is
necessary to weed out irrelevant and incorrect pairings to extract the concept
lattice and hierarchies more quickly. This study aims to propose a framework
for reducing formal context in extracting concept hierarchies from free text to
reduce the ambiguity of the formal context. We achieve this by reducing the
size of the formal context using a hybrid of a WordNet-based method and a
frequency-based technique. Using 385 samples from the Wikipedia corpus and the
suggested framework, tests are carried out to examine the reduced size of
formal context, leading to concept lattice and concept hierarchy. With the help
of concept lattice-invariants, the generated formal context lattice is compared
to the normal one. In contrast to basic ones, the homomorphic between the
resultant lattices retains up to 98% of the quality of the generating concept
hierarchies, and the reduced concept lattice receives the structural connection
of the standard one. Additionally, the new framework is compared to five
baseline techniques to calculate the running time on random datasets with
various densities. The findings demonstrate that, in various fill ratios,
hybrid approaches of the proposed method outperform other indicated competing
strategies in concept lattice performance.
|
2504.06292 | Hezhe Qiao | Hongbin Liang, Hezhe Qiao, Wei Huang, Qizhou Wang, Mingsheng Shang,
and Lin Chen | Temporal-contextual Event Learning for Pedestrian Crossing Intent
Prediction | Accepted in ICONIP2024 | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Ensuring the safety of vulnerable road users through accurate prediction of
pedestrian crossing intention (PCI) plays a crucial role in the context of
autonomous and assisted driving. Analyzing the set of observation video frames
in ego-view has been widely used in most PCI prediction methods to forecast the
cross intent. However, they struggle to capture the critical events related to
pedestrian behaviour along the temporal dimension due to the high redundancy of
the video frames, which results in the sub-optimal performance of PCI
prediction. Our research addresses the challenge by introducing a novel
approach called \underline{T}emporal-\underline{c}ontextual Event
\underline{L}earning (TCL). The TCL is composed of the Temporal Merging Module
(TMM), which aims to manage the redundancy by clustering the observed video
frames into multiple key temporal events. Then, the Contextual Attention Block
(CAB) is employed to adaptively aggregate multiple event features along with
visual and non-visual data. By synthesizing the temporal feature extraction and
contextual attention on the key information across the critical events, TCL can
learn expressive representation for the PCI prediction. Extensive experiments
are carried out on three widely adopted datasets, including PIE, JAAD-beh, and
JAAD-all. The results show that TCL substantially surpasses the
state-of-the-art methods. Our code can be accessed at
https://github.com/dadaguailhb/TCL.
| [
{
"version": "v1",
"created": "Fri, 4 Apr 2025 10:44:24 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Liang",
"Hongbin",
""
],
[
"Qiao",
"Hezhe",
""
],
[
"Huang",
"Wei",
""
],
[
"Wang",
"Qizhou",
""
],
[
"Shang",
"Mingsheng",
""
],
[
"Chen",
"Lin",
""
]
] | TITLE: Temporal-contextual Event Learning for Pedestrian Crossing Intent
Prediction
ABSTRACT: Ensuring the safety of vulnerable road users through accurate prediction of
pedestrian crossing intention (PCI) plays a crucial role in the context of
autonomous and assisted driving. Analyzing the set of observation video frames
in ego-view has been widely used in most PCI prediction methods to forecast the
cross intent. However, they struggle to capture the critical events related to
pedestrian behaviour along the temporal dimension due to the high redundancy of
the video frames, which results in the sub-optimal performance of PCI
prediction. Our research addresses the challenge by introducing a novel
approach called \underline{T}emporal-\underline{c}ontextual Event
\underline{L}earning (TCL). The TCL is composed of the Temporal Merging Module
(TMM), which aims to manage the redundancy by clustering the observed video
frames into multiple key temporal events. Then, the Contextual Attention Block
(CAB) is employed to adaptively aggregate multiple event features along with
visual and non-visual data. By synthesizing the temporal feature extraction and
contextual attention on the key information across the critical events, TCL can
learn expressive representation for the PCI prediction. Extensive experiments
are carried out on three widely adopted datasets, including PIE, JAAD-beh, and
JAAD-all. The results show that TCL substantially surpasses the
state-of-the-art methods. Our code can be accessed at
https://github.com/dadaguailhb/TCL.
|
2504.06306 | Polycarp Nalela | Polycarp Nalela, Deepthi Rao, Praveen Rao | Predicting Survivability of Cancer Patients with Metastatic Patterns
Using Explainable AI | null | null | null | null | q-bio.QM cs.AI | http://creativecommons.org/licenses/by/4.0/ | Cancer remains a leading global health challenge and a major cause of
mortality. This study leverages machine learning (ML) to predict the
survivability of cancer patients with metastatic patterns using the
comprehensive MSK-MET dataset, which includes genomic and clinical data from
25,775 patients across 27 cancer types. We evaluated five ML models-XGBoost,
Na\"ive Bayes, Decision Tree, Logistic Regression, and Random Fores using
hyperparameter tuning and grid search. XGBoost emerged as the best performer
with an area under the curve (AUC) of 0.82. To enhance model interpretability,
SHapley Additive exPlanations (SHAP) were applied, revealing key predictors
such as metastatic site count, tumor mutation burden, fraction of genome
altered, and organ-specific metastases. Further survival analysis using
Kaplan-Meier curves, Cox Proportional Hazards models, and XGBoost Survival
Analysis identified significant predictors of patient outcomes, offering
actionable insights for clinicians. These findings could aid in personalized
prognosis and treatment planning, ultimately improving patient care.
| [
{
"version": "v1",
"created": "Mon, 7 Apr 2025 20:48:15 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Nalela",
"Polycarp",
""
],
[
"Rao",
"Deepthi",
""
],
[
"Rao",
"Praveen",
""
]
] | TITLE: Predicting Survivability of Cancer Patients with Metastatic Patterns
Using Explainable AI
ABSTRACT: Cancer remains a leading global health challenge and a major cause of
mortality. This study leverages machine learning (ML) to predict the
survivability of cancer patients with metastatic patterns using the
comprehensive MSK-MET dataset, which includes genomic and clinical data from
25,775 patients across 27 cancer types. We evaluated five ML models-XGBoost,
Na\"ive Bayes, Decision Tree, Logistic Regression, and Random Fores using
hyperparameter tuning and grid search. XGBoost emerged as the best performer
with an area under the curve (AUC) of 0.82. To enhance model interpretability,
SHapley Additive exPlanations (SHAP) were applied, revealing key predictors
such as metastatic site count, tumor mutation burden, fraction of genome
altered, and organ-specific metastases. Further survival analysis using
Kaplan-Meier curves, Cox Proportional Hazards models, and XGBoost Survival
Analysis identified significant predictors of patient outcomes, offering
actionable insights for clinicians. These findings could aid in personalized
prognosis and treatment planning, ultimately improving patient care.
|
2504.06312 | Peizhi Niu | Peizhi Niu, Yu-Hsiang Wang, Vishal Rana, Chetan Rupakheti, Abhishek
Pandey, Olgica Milenkovic | DMol: A Schedule-Driven Diffusion Model for Highly Efficient and
Versatile Molecule Generation | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce a new graph diffusion model for small molecule generation,
\emph{DMol}, which outperforms the state-of-the-art DiGress model in terms of
validity by roughly $1.5\%$ across all benchmarking datasets while reducing the
number of diffusion steps by at least $10$-fold, and the running time to
roughly one half. The performance improvements are a result of a careful change
in the objective function and a ``graph noise" scheduling approach which, at
each diffusion step, allows one to only change a subset of nodes of varying
size in the molecule graph. Another relevant property of the method is that it
can be easily combined with junction-tree-like graph representations that arise
by compressing a collection of relevant ring structures into supernodes. Unlike
classical junction-tree techniques that involve VAEs and require complicated
reconstruction steps, compressed DMol directly performs graph diffusion on a
graph that compresses only a carefully selected set of frequent carbon rings
into supernodes, which results in straightforward sample generation. This
compressed DMol method offers additional validity improvements over generic
DMol of roughly $2\%$, increases the novelty of the method, and further
improves the running time due to reductions in the graph size.
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 03:31:21 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Niu",
"Peizhi",
""
],
[
"Wang",
"Yu-Hsiang",
""
],
[
"Rana",
"Vishal",
""
],
[
"Rupakheti",
"Chetan",
""
],
[
"Pandey",
"Abhishek",
""
],
[
"Milenkovic",
"Olgica",
""
]
] | TITLE: DMol: A Schedule-Driven Diffusion Model for Highly Efficient and
Versatile Molecule Generation
ABSTRACT: We introduce a new graph diffusion model for small molecule generation,
\emph{DMol}, which outperforms the state-of-the-art DiGress model in terms of
validity by roughly $1.5\%$ across all benchmarking datasets while reducing the
number of diffusion steps by at least $10$-fold, and the running time to
roughly one half. The performance improvements are a result of a careful change
in the objective function and a ``graph noise" scheduling approach which, at
each diffusion step, allows one to only change a subset of nodes of varying
size in the molecule graph. Another relevant property of the method is that it
can be easily combined with junction-tree-like graph representations that arise
by compressing a collection of relevant ring structures into supernodes. Unlike
classical junction-tree techniques that involve VAEs and require complicated
reconstruction steps, compressed DMol directly performs graph diffusion on a
graph that compresses only a carefully selected set of frequent carbon rings
into supernodes, which results in straightforward sample generation. This
compressed DMol method offers additional validity improvements over generic
DMol of roughly $2\%$, increases the novelty of the method, and further
improves the running time due to reductions in the graph size.
|
2504.06314 | Abdelghani MADDI | Abdelghani Maddi (GEMASS), Jaime Teixeira Da Silva (MIDAP) | Beyond authorship: Analyzing contributions in PLOS ONE and the
challenges of appropriate attribution | null | Journal of Data and Information Science, 2024, 9 (3), pp.88-115 | 10.2478/jdis-2024-0015 | null | cs.DL cs.CY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Purpose This study aims to evaluate the accuracy of authorship attributions
in scientific publications, focusing on the fairness and precision of
individual contributions within academic works. Design/methodology/approach The
study analyzes 81,823 publications from the journal PLOS ONE , covering the
period from January 2018 to June 2023. It examines the authorship attributions
within these publications to try and determine the prevalence of inappropriate
authorship. It also investigates the demographic and professional profiles of
affected authors, exploring trends and potential factors contributing to
inaccuracies in authorship. Findings Surprisingly, 9.14% of articles feature at
least one author with inappropriate authorship, affecting over 14,000
individuals (2.56% of the sample). Inappropriate authorship is more
concentrated in Asia, Africa, and specific European countries like Italy.
Established researchers with significant publication records and those
affiliated with companies or nonprofits show higher instances of potential
monetary authorship. Research limitations Our findings are based on
contributions as declared by the authors, which implies a degree of trust in
their transparency. However, this reliance on self-reporting may introduce
biases or inaccuracies into the dataset. Further research could employ
additional verification methods to enhance the reliability of the findings.
Practical implications These findings have significant implications for journal
publishers, highlighting the necessity for robust control mechanisms to ensure
the integrity of authorship attributions. Moreover, researchers must exercise
discernment in determining when to acknowledge a contributor and when to
include them in the author list. Addressing these issues is crucial for
maintaining the credibility and fairness of academic publications.
Originality/value This study contributes to an understanding of critical issues
within academic authorship, shedding light on the prevalence and impact of
inappropriate authorship attributions. By calling for a nuanced approach to
ensure accurate credit is given where it is due, the study underscores the
importance of upholding ethical standards in scholarly publishing.
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 06:47:52 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Maddi",
"Abdelghani",
"",
"GEMASS"
],
[
"Da Silva",
"Jaime Teixeira",
"",
"MIDAP"
]
] | TITLE: Beyond authorship: Analyzing contributions in PLOS ONE and the
challenges of appropriate attribution
ABSTRACT: Purpose This study aims to evaluate the accuracy of authorship attributions
in scientific publications, focusing on the fairness and precision of
individual contributions within academic works. Design/methodology/approach The
study analyzes 81,823 publications from the journal PLOS ONE , covering the
period from January 2018 to June 2023. It examines the authorship attributions
within these publications to try and determine the prevalence of inappropriate
authorship. It also investigates the demographic and professional profiles of
affected authors, exploring trends and potential factors contributing to
inaccuracies in authorship. Findings Surprisingly, 9.14% of articles feature at
least one author with inappropriate authorship, affecting over 14,000
individuals (2.56% of the sample). Inappropriate authorship is more
concentrated in Asia, Africa, and specific European countries like Italy.
Established researchers with significant publication records and those
affiliated with companies or nonprofits show higher instances of potential
monetary authorship. Research limitations Our findings are based on
contributions as declared by the authors, which implies a degree of trust in
their transparency. However, this reliance on self-reporting may introduce
biases or inaccuracies into the dataset. Further research could employ
additional verification methods to enhance the reliability of the findings.
Practical implications These findings have significant implications for journal
publishers, highlighting the necessity for robust control mechanisms to ensure
the integrity of authorship attributions. Moreover, researchers must exercise
discernment in determining when to acknowledge a contributor and when to
include them in the author list. Addressing these issues is crucial for
maintaining the credibility and fairness of academic publications.
Originality/value This study contributes to an understanding of critical issues
within academic authorship, shedding light on the prevalence and impact of
inappropriate authorship attributions. By calling for a nuanced approach to
ensure accurate credit is given where it is due, the study underscores the
importance of upholding ethical standards in scholarly publishing.
|
2504.06318 | Mathias Angermaier | Mathias Angermaier and Joao Pinheiro-Neto and Elisabeth Hoeldrich and
Jana Lasser | The Schwurbelarchiv: a German Language Telegram dataset for the Study of
Conspiracy Theories | null | null | null | null | cs.SI | http://creativecommons.org/licenses/by/4.0/ | Sociality borne by language, as is the predominant digital trace on
text-based social media platforms, harbours the raw material for exploring
multiple social phenomena. Distinctively, the messaging service Telegram
provides functionalities that allow for socially interactive as well as
one-to-many communication. Our Telegram dataset contains over 6,000 groups and
channels, 40 million text messages, and over 3 million transcribed audio files,
originating from a data-hoarding initiative named the ``Schwurbelarchiv'' (from
German schwurbeln: speaking nonsense). This dataset publication details the
structure, scope, and methodological specifics of the Schwurbelarchiv,
emphasising its relevance for further research on the German-language
conspiracy theory discourse. We validate its predominantly German origin by
linguistic and temporal markers and situate it within the context of similar
datasets. We describe process and extent of the transcription of multimedia
files. Thanks to this effort the dataset uniquely supports multimodal analysis
of online social dynamics and content dissemination. Researchers can employ
this resource to explore societal dynamics in misinformation, political
extremism, opinion adaptation, and social network structures on Telegram. The
Schwurbelarchiv thus offers unprecedented opportunities for investigations into
digital communication and its societal implications.
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 09:11:46 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Angermaier",
"Mathias",
""
],
[
"Pinheiro-Neto",
"Joao",
""
],
[
"Hoeldrich",
"Elisabeth",
""
],
[
"Lasser",
"Jana",
""
]
] | TITLE: The Schwurbelarchiv: a German Language Telegram dataset for the Study of
Conspiracy Theories
ABSTRACT: Sociality borne by language, as is the predominant digital trace on
text-based social media platforms, harbours the raw material for exploring
multiple social phenomena. Distinctively, the messaging service Telegram
provides functionalities that allow for socially interactive as well as
one-to-many communication. Our Telegram dataset contains over 6,000 groups and
channels, 40 million text messages, and over 3 million transcribed audio files,
originating from a data-hoarding initiative named the ``Schwurbelarchiv'' (from
German schwurbeln: speaking nonsense). This dataset publication details the
structure, scope, and methodological specifics of the Schwurbelarchiv,
emphasising its relevance for further research on the German-language
conspiracy theory discourse. We validate its predominantly German origin by
linguistic and temporal markers and situate it within the context of similar
datasets. We describe process and extent of the transcription of multimedia
files. Thanks to this effort the dataset uniquely supports multimodal analysis
of online social dynamics and content dissemination. Researchers can employ
this resource to explore societal dynamics in misinformation, political
extremism, opinion adaptation, and social network structures on Telegram. The
Schwurbelarchiv thus offers unprecedented opportunities for investigations into
digital communication and its societal implications.
|
2504.06323 | Bailey Eccles | Bailey J. Eccles, Leon Wong, Blesson Varghese | Mosaic: Composite Projection Pruning for Resource-efficient LLMs | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Extensive compute and memory requirements limit the deployment of large
language models (LLMs) on any hardware. Compression methods, such as pruning,
can reduce model size, which in turn reduces resource requirements.
State-of-the-art pruning is based on coarse-grained methods. They are
time-consuming and inherently remove critical model parameters, adversely
impacting the quality of the pruned model. This paper introduces projection
pruning, a novel fine-grained method for pruning LLMs. In addition, LLM
projection pruning is enhanced by a new approach we refer to as composite
projection pruning - the synergistic combination of unstructured pruning that
retains accuracy and structured pruning that reduces model size. We develop
Mosaic, a novel system to create and deploy pruned LLMs using composite
projection pruning. Mosaic is evaluated using a range of performance and
quality metrics on multiple hardware platforms, LLMs, and datasets. Mosaic is
7.19x faster in producing models than existing approaches. Mosaic models
achieve up to 84.2% lower perplexity and 31.4% higher accuracy than models
obtained from coarse-grained pruning. Up to 67% faster inference and 68% lower
GPU memory use is noted for Mosaic models.
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 11:51:35 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Eccles",
"Bailey J.",
""
],
[
"Wong",
"Leon",
""
],
[
"Varghese",
"Blesson",
""
]
] | TITLE: Mosaic: Composite Projection Pruning for Resource-efficient LLMs
ABSTRACT: Extensive compute and memory requirements limit the deployment of large
language models (LLMs) on any hardware. Compression methods, such as pruning,
can reduce model size, which in turn reduces resource requirements.
State-of-the-art pruning is based on coarse-grained methods. They are
time-consuming and inherently remove critical model parameters, adversely
impacting the quality of the pruned model. This paper introduces projection
pruning, a novel fine-grained method for pruning LLMs. In addition, LLM
projection pruning is enhanced by a new approach we refer to as composite
projection pruning - the synergistic combination of unstructured pruning that
retains accuracy and structured pruning that reduces model size. We develop
Mosaic, a novel system to create and deploy pruned LLMs using composite
projection pruning. Mosaic is evaluated using a range of performance and
quality metrics on multiple hardware platforms, LLMs, and datasets. Mosaic is
7.19x faster in producing models than existing approaches. Mosaic models
achieve up to 84.2% lower perplexity and 31.4% higher accuracy than models
obtained from coarse-grained pruning. Up to 67% faster inference and 68% lower
GPU memory use is noted for Mosaic models.
|
2504.06324 | Monika Jotautait\.e | Monika Jotautaite, Mary Phuong, Chatrik Singh Mangat, Maria Angelica
Martinez | From Stability to Inconsistency: A Study of Moral Preferences in LLMs | null | null | null | null | cs.CY cs.AI | http://creativecommons.org/licenses/by/4.0/ | As large language models (LLMs) increasingly integrate into our daily lives,
it becomes crucial to understand their implicit biases and moral tendencies. To
address this, we introduce a Moral Foundations LLM dataset (MFD-LLM) grounded
in Moral Foundations Theory, which conceptualizes human morality through six
core foundations. We propose a novel evaluation method that captures the full
spectrum of LLMs' revealed moral preferences by answering a range of real-world
moral dilemmas. Our findings reveal that state-of-the-art models have
remarkably homogeneous value preferences, yet demonstrate a lack of
consistency.
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 11:52:50 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Jotautaite",
"Monika",
""
],
[
"Phuong",
"Mary",
""
],
[
"Mangat",
"Chatrik Singh",
""
],
[
"Martinez",
"Maria Angelica",
""
]
] | TITLE: From Stability to Inconsistency: A Study of Moral Preferences in LLMs
ABSTRACT: As large language models (LLMs) increasingly integrate into our daily lives,
it becomes crucial to understand their implicit biases and moral tendencies. To
address this, we introduce a Moral Foundations LLM dataset (MFD-LLM) grounded
in Moral Foundations Theory, which conceptualizes human morality through six
core foundations. We propose a novel evaluation method that captures the full
spectrum of LLMs' revealed moral preferences by answering a range of real-world
moral dilemmas. Our findings reveal that state-of-the-art models have
remarkably homogeneous value preferences, yet demonstrate a lack of
consistency.
|
2504.06325 | Wenbin Xing | Ronghui Zhang, Wenbin Xing, Mengran Li, Zihan Wang, Junzhou Chen,
Xiaolei Ma, Zhiyuan Liu, Zhengbing He | MM-STFlowNet: A Transportation Hub-Oriented Multi-Mode Passenger Flow
Prediction Method via Spatial-Temporal Dynamic Graph Modeling | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Accurate and refined passenger flow prediction is essential for optimizing
the collaborative management of multiple collection and distribution modes in
large-scale transportation hubs. Traditional methods often focus only on the
overall passenger volume, neglecting the interdependence between different
modes within the hub. To address this limitation, we propose MM-STFlowNet, a
comprehensive multi-mode prediction framework grounded in dynamic
spatial-temporal graph modeling. Initially, an integrated temporal feature
processing strategy is implemented using signal decomposition and convolution
techniques to address data spikes and high volatility. Subsequently, we
introduce the Spatial-Temporal Dynamic Graph Convolutional Recurrent Network
(STDGCRN) to capture detailed spatial-temporal dependencies across multiple
traffic modes, enhanced by an adaptive channel attention mechanism. Finally,
the self-attention mechanism is applied to incorporate various external
factors, further enhancing prediction accuracy. Experiments on a real-world
dataset from Guangzhounan Railway Station in China demonstrate that
MM-STFlowNet achieves state-of-the-art performance, particularly during peak
periods, providing valuable insight for transportation hub management.
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 12:00:06 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Zhang",
"Ronghui",
""
],
[
"Xing",
"Wenbin",
""
],
[
"Li",
"Mengran",
""
],
[
"Wang",
"Zihan",
""
],
[
"Chen",
"Junzhou",
""
],
[
"Ma",
"Xiaolei",
""
],
[
"Liu",
"Zhiyuan",
""
],
[
"He",
"Zhengbing",
""
]
] | TITLE: MM-STFlowNet: A Transportation Hub-Oriented Multi-Mode Passenger Flow
Prediction Method via Spatial-Temporal Dynamic Graph Modeling
ABSTRACT: Accurate and refined passenger flow prediction is essential for optimizing
the collaborative management of multiple collection and distribution modes in
large-scale transportation hubs. Traditional methods often focus only on the
overall passenger volume, neglecting the interdependence between different
modes within the hub. To address this limitation, we propose MM-STFlowNet, a
comprehensive multi-mode prediction framework grounded in dynamic
spatial-temporal graph modeling. Initially, an integrated temporal feature
processing strategy is implemented using signal decomposition and convolution
techniques to address data spikes and high volatility. Subsequently, we
introduce the Spatial-Temporal Dynamic Graph Convolutional Recurrent Network
(STDGCRN) to capture detailed spatial-temporal dependencies across multiple
traffic modes, enhanced by an adaptive channel attention mechanism. Finally,
the self-attention mechanism is applied to incorporate various external
factors, further enhancing prediction accuracy. Experiments on a real-world
dataset from Guangzhounan Railway Station in China demonstrate that
MM-STFlowNet achieves state-of-the-art performance, particularly during peak
periods, providing valuable insight for transportation hub management.
|
2504.06327 | Ali Kashefi | Ali Kashefi, Tapan Mukerji | Physics-informed KAN PointNet: Deep learning for simultaneous solutions
to inverse problems in incompressible flow on numerous irregular geometries | null | null | null | null | cs.LG physics.flu-dyn | http://creativecommons.org/licenses/by/4.0/ | Kolmogorov-Arnold Networks (KANs) have gained attention as a promising
alternative to traditional Multilayer Perceptrons (MLPs) for deep learning
applications in computational physics, especially within the framework of
physics-informed neural networks (PINNs). Physics-informed Kolmogorov-Arnold
Networks (PIKANs) and their variants have been introduced and evaluated to
solve inverse problems. However, similar to PINNs, current versions of PIKANs
are limited to obtaining solutions for a single computational domain per
training run; consequently, a new geometry requires retraining the model from
scratch. Physics-informed PointNet (PIPN) was introduced to address this
limitation for PINNs. In this work, we introduce physics-informed
Kolmogorov-Arnold PointNet (PI-KAN-PointNet) to extend this capability to
PIKANs. PI-KAN-PointNet enables the simultaneous solution of an inverse problem
over multiple irregular geometries within a single training run, reducing
computational costs. We construct KANs using Jacobi polynomials and investigate
their performance by considering Jacobi polynomials of different degrees and
types in terms of both computational cost and prediction accuracy. As a
benchmark test case, we consider natural convection in a square enclosure with
a cylinder, where the cylinder's shape varies across a dataset of 135
geometries. We compare the performance of PI-KAN-PointNet with that of PIPN
(i.e., physics-informed PointNet with MLPs) and observe that, with
approximately an equal number of trainable parameters and similar computational
cost, PI-KAN-PointNet provides more accurate predictions. Finally, we explore
the combination of KAN and MLP in constructing a physics-informed PointNet. Our
findings indicate that a physics-informed PointNet model employing MLP layers
as the encoder and KAN layers as the decoder represents the optimal
configuration among all models investigated.
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 12:31:57 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Kashefi",
"Ali",
""
],
[
"Mukerji",
"Tapan",
""
]
] | TITLE: Physics-informed KAN PointNet: Deep learning for simultaneous solutions
to inverse problems in incompressible flow on numerous irregular geometries
ABSTRACT: Kolmogorov-Arnold Networks (KANs) have gained attention as a promising
alternative to traditional Multilayer Perceptrons (MLPs) for deep learning
applications in computational physics, especially within the framework of
physics-informed neural networks (PINNs). Physics-informed Kolmogorov-Arnold
Networks (PIKANs) and their variants have been introduced and evaluated to
solve inverse problems. However, similar to PINNs, current versions of PIKANs
are limited to obtaining solutions for a single computational domain per
training run; consequently, a new geometry requires retraining the model from
scratch. Physics-informed PointNet (PIPN) was introduced to address this
limitation for PINNs. In this work, we introduce physics-informed
Kolmogorov-Arnold PointNet (PI-KAN-PointNet) to extend this capability to
PIKANs. PI-KAN-PointNet enables the simultaneous solution of an inverse problem
over multiple irregular geometries within a single training run, reducing
computational costs. We construct KANs using Jacobi polynomials and investigate
their performance by considering Jacobi polynomials of different degrees and
types in terms of both computational cost and prediction accuracy. As a
benchmark test case, we consider natural convection in a square enclosure with
a cylinder, where the cylinder's shape varies across a dataset of 135
geometries. We compare the performance of PI-KAN-PointNet with that of PIPN
(i.e., physics-informed PointNet with MLPs) and observe that, with
approximately an equal number of trainable parameters and similar computational
cost, PI-KAN-PointNet provides more accurate predictions. Finally, we explore
the combination of KAN and MLP in constructing a physics-informed PointNet. Our
findings indicate that a physics-informed PointNet model employing MLP layers
as the encoder and KAN layers as the decoder represents the optimal
configuration among all models investigated.
|
2504.06330 | Hicham Talaoubrid | Hicham Talaoubrid, Anissa Mokraoui, Ismail Ben Ayed, Axel Prouvost,
Sonimith Hang, Monit Korn, R\'emi Harvey | Analyzing the Impact of Low-Rank Adaptation for Cross-Domain Few-Shot
Object Detection in Aerial Images | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | This paper investigates the application of Low-Rank Adaptation (LoRA) to
small models for cross-domain few-shot object detection in aerial images.
Originally designed for large-scale models, LoRA helps mitigate overfitting,
making it a promising approach for resource-constrained settings. We integrate
LoRA into DiffusionDet, and evaluate its performance on the DOTA and DIOR
datasets. Our results show that LoRA applied after an initial fine-tuning
slightly improves performance in low-shot settings (e.g., 1-shot and 5-shot),
while full fine-tuning remains more effective in higher-shot configurations.
These findings highlight LoRA's potential for efficient adaptation in aerial
object detection, encouraging further research into parameter-efficient
fine-tuning strategies for few-shot learning. Our code is available here:
https://github.com/HichTala/LoRA-DiffusionDet.
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 14:10:39 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Talaoubrid",
"Hicham",
""
],
[
"Mokraoui",
"Anissa",
""
],
[
"Ayed",
"Ismail Ben",
""
],
[
"Prouvost",
"Axel",
""
],
[
"Hang",
"Sonimith",
""
],
[
"Korn",
"Monit",
""
],
[
"Harvey",
"Rémi",
""
]
] | TITLE: Analyzing the Impact of Low-Rank Adaptation for Cross-Domain Few-Shot
Object Detection in Aerial Images
ABSTRACT: This paper investigates the application of Low-Rank Adaptation (LoRA) to
small models for cross-domain few-shot object detection in aerial images.
Originally designed for large-scale models, LoRA helps mitigate overfitting,
making it a promising approach for resource-constrained settings. We integrate
LoRA into DiffusionDet, and evaluate its performance on the DOTA and DIOR
datasets. Our results show that LoRA applied after an initial fine-tuning
slightly improves performance in low-shot settings (e.g., 1-shot and 5-shot),
while full fine-tuning remains more effective in higher-shot configurations.
These findings highlight LoRA's potential for efficient adaptation in aerial
object detection, encouraging further research into parameter-efficient
fine-tuning strategies for few-shot learning. Our code is available here:
https://github.com/HichTala/LoRA-DiffusionDet.
|
2504.06358 | Yupeng Cheng | Yupeng Cheng, Zi Pong Lim, Sarthak Ketanbhai Modi, Yon Shin Teo, Yushi
Cao, Shang-Wei Lin | Towards Calibration Enhanced Network by Inverse Adversarial Attack | 11 pages | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Test automation has become increasingly important as the complexity of both
design and content in Human Machine Interface (HMI) software continues to grow.
Current standard practice uses Optical Character Recognition (OCR) techniques
to automatically extract textual information from HMI screens for validation.
At present, one of the key challenges faced during the automation of HMI screen
validation is the noise handling for the OCR models. In this paper, we propose
to utilize adversarial training techniques to enhance OCR models in HMI testing
scenarios. More specifically, we design a new adversarial attack objective for
OCR models to discover the decision boundaries in the context of HMI testing.
We then adopt adversarial training to optimize the decision boundaries towards
a more robust and accurate OCR model. In addition, we also built an HMI screen
dataset based on real-world requirements and applied multiple types of
perturbation onto the clean HMI dataset to provide a more complete coverage for
the potential scenarios. We conduct experiments to demonstrate how using
adversarial training techniques yields more robust OCR models against various
kinds of noises, while still maintaining high OCR model accuracy. Further
experiments even demonstrate that the adversarial training models exhibit a
certain degree of robustness against perturbations from other patterns.
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 18:13:23 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Cheng",
"Yupeng",
""
],
[
"Lim",
"Zi Pong",
""
],
[
"Modi",
"Sarthak Ketanbhai",
""
],
[
"Teo",
"Yon Shin",
""
],
[
"Cao",
"Yushi",
""
],
[
"Lin",
"Shang-Wei",
""
]
] | TITLE: Towards Calibration Enhanced Network by Inverse Adversarial Attack
ABSTRACT: Test automation has become increasingly important as the complexity of both
design and content in Human Machine Interface (HMI) software continues to grow.
Current standard practice uses Optical Character Recognition (OCR) techniques
to automatically extract textual information from HMI screens for validation.
At present, one of the key challenges faced during the automation of HMI screen
validation is the noise handling for the OCR models. In this paper, we propose
to utilize adversarial training techniques to enhance OCR models in HMI testing
scenarios. More specifically, we design a new adversarial attack objective for
OCR models to discover the decision boundaries in the context of HMI testing.
We then adopt adversarial training to optimize the decision boundaries towards
a more robust and accurate OCR model. In addition, we also built an HMI screen
dataset based on real-world requirements and applied multiple types of
perturbation onto the clean HMI dataset to provide a more complete coverage for
the potential scenarios. We conduct experiments to demonstrate how using
adversarial training techniques yields more robust OCR models against various
kinds of noises, while still maintaining high OCR model accuracy. Further
experiments even demonstrate that the adversarial training models exhibit a
certain degree of robustness against perturbations from other patterns.
|
2504.06393 | Rebecca M. M. Hicke | Rebecca M. M. Hicke, Sil Hamilton, and David Mimno | The Zero Body Problem: Probing LLM Use of Sensory Language | null | null | null | null | cs.CL cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Sensory language expresses embodied experiences ranging from taste and sound
to excitement and stomachache. This language is of interest to scholars from a
wide range of domains including robotics, narratology, linguistics, and
cognitive science. In this work, we explore whether language models, which are
not embodied, can approximate human use of embodied language. We extend an
existing corpus of parallel human and model responses to short story prompts
with an additional 18,000 stories generated by 18 popular models. We find that
all models generate stories that differ significantly from human usage of
sensory language, but the direction of these differences varies considerably
between model families. Namely, Gemini models use significantly more sensory
language than humans along most axes whereas most models from the remaining
five families use significantly less. Linear probes run on five models suggest
that they are capable of identifying sensory language. However, we find
preliminary evidence suggesting that instruction tuning may discourage usage of
sensory language. Finally, to support further work, we release our expanded
story dataset.
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 19:31:37 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Hicke",
"Rebecca M. M.",
""
],
[
"Hamilton",
"Sil",
""
],
[
"Mimno",
"David",
""
]
] | TITLE: The Zero Body Problem: Probing LLM Use of Sensory Language
ABSTRACT: Sensory language expresses embodied experiences ranging from taste and sound
to excitement and stomachache. This language is of interest to scholars from a
wide range of domains including robotics, narratology, linguistics, and
cognitive science. In this work, we explore whether language models, which are
not embodied, can approximate human use of embodied language. We extend an
existing corpus of parallel human and model responses to short story prompts
with an additional 18,000 stories generated by 18 popular models. We find that
all models generate stories that differ significantly from human usage of
sensory language, but the direction of these differences varies considerably
between model families. Namely, Gemini models use significantly more sensory
language than humans along most axes whereas most models from the remaining
five families use significantly less. Linear probes run on five models suggest
that they are capable of identifying sensory language. However, we find
preliminary evidence suggesting that instruction tuning may discourage usage of
sensory language. Finally, to support further work, we release our expanded
story dataset.
|
2504.06410 | Huzaifa Arif | Huzaifa Arif, Keerthiram Murugesan, Payel Das, Alex Gittens, Pin-Yu
Chen | PEEL the Layers and Find Yourself: Revisiting Inference-time Data
Leakage for Residual Neural Networks | null | null | null | null | cs.LG cs.CR cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper explores inference-time data leakage risks of deep neural networks
(NNs), where a curious and honest model service provider is interested in
retrieving users' private data inputs solely based on the model inference
results. Particularly, we revisit residual NNs due to their popularity in
computer vision and our hypothesis that residual blocks are a primary cause of
data leakage owing to the use of skip connections. By formulating
inference-time data leakage as a constrained optimization problem, we propose a
novel backward feature inversion method, \textbf{PEEL}, which can effectively
recover block-wise input features from the intermediate output of residual NNs.
The surprising results in high-quality input data recovery can be explained by
the intuition that the output from these residual blocks can be considered as a
noisy version of the input and thus the output retains sufficient information
for input recovery. We demonstrate the effectiveness of our layer-by-layer
feature inversion method on facial image datasets and pre-trained classifiers.
Our results show that PEEL outperforms the state-of-the-art recovery methods by
an order of magnitude when evaluated by mean squared error (MSE). The code is
available at
\href{https://github.com/Huzaifa-Arif/PEEL}{https://github.com/Huzaifa-Arif/PEEL}
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 20:11:05 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Arif",
"Huzaifa",
""
],
[
"Murugesan",
"Keerthiram",
""
],
[
"Das",
"Payel",
""
],
[
"Gittens",
"Alex",
""
],
[
"Chen",
"Pin-Yu",
""
]
] | TITLE: PEEL the Layers and Find Yourself: Revisiting Inference-time Data
Leakage for Residual Neural Networks
ABSTRACT: This paper explores inference-time data leakage risks of deep neural networks
(NNs), where a curious and honest model service provider is interested in
retrieving users' private data inputs solely based on the model inference
results. Particularly, we revisit residual NNs due to their popularity in
computer vision and our hypothesis that residual blocks are a primary cause of
data leakage owing to the use of skip connections. By formulating
inference-time data leakage as a constrained optimization problem, we propose a
novel backward feature inversion method, \textbf{PEEL}, which can effectively
recover block-wise input features from the intermediate output of residual NNs.
The surprising results in high-quality input data recovery can be explained by
the intuition that the output from these residual blocks can be considered as a
noisy version of the input and thus the output retains sufficient information
for input recovery. We demonstrate the effectiveness of our layer-by-layer
feature inversion method on facial image datasets and pre-trained classifiers.
Our results show that PEEL outperforms the state-of-the-art recovery methods by
an order of magnitude when evaluated by mean squared error (MSE). The code is
available at
\href{https://github.com/Huzaifa-Arif/PEEL}{https://github.com/Huzaifa-Arif/PEEL}
|
2504.06417 | Ildi Alla | Ildi Alla, Selma Yahia, Valeria Loscri | TRIDENT: Tri-modal Real-time Intrusion Detection Engine for New Targets | null | null | null | null | cs.CR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The increasing availability of drones and their potential for malicious
activities pose significant privacy and security risks, necessitating fast and
reliable detection in real-world environments. However, existing drone
detection systems often struggle in real-world settings due to environmental
noise and sensor limitations. This paper introduces TRIDENT, a tri-modal drone
detection framework that integrates synchronized audio, visual, and RF data to
enhance robustness and reduce dependence on individual sensors. TRIDENT
introduces two fusion strategies - Late Fusion and GMU Fusion - to improve
multi-modal integration while maintaining efficiency. The framework
incorporates domain-specific feature extraction techniques alongside a
specialized data augmentation pipeline that simulates real-world sensor
degradation to improve generalization capabilities. A diverse multi-sensor
dataset is collected in urban and non-urban environments under varying lighting
conditions, ensuring comprehensive evaluation. Experimental results show that
TRIDENT achieves 98.8 percent accuracy in real-world recordings and 83.26
percent in a more complex setting (augmented data), outperforming unimodal and
dual-modal baselines. Moreover, TRIDENT operates in real-time, detecting drones
in just 6.09 ms while consuming only 75.27 mJ per detection, making it highly
efficient for resource-constrained devices. The dataset and code have been
released to ensure reproducibility (https://github.com/TRIDENT-2025/TRIDENT).
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 20:33:43 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Alla",
"Ildi",
""
],
[
"Yahia",
"Selma",
""
],
[
"Loscri",
"Valeria",
""
]
] | TITLE: TRIDENT: Tri-modal Real-time Intrusion Detection Engine for New Targets
ABSTRACT: The increasing availability of drones and their potential for malicious
activities pose significant privacy and security risks, necessitating fast and
reliable detection in real-world environments. However, existing drone
detection systems often struggle in real-world settings due to environmental
noise and sensor limitations. This paper introduces TRIDENT, a tri-modal drone
detection framework that integrates synchronized audio, visual, and RF data to
enhance robustness and reduce dependence on individual sensors. TRIDENT
introduces two fusion strategies - Late Fusion and GMU Fusion - to improve
multi-modal integration while maintaining efficiency. The framework
incorporates domain-specific feature extraction techniques alongside a
specialized data augmentation pipeline that simulates real-world sensor
degradation to improve generalization capabilities. A diverse multi-sensor
dataset is collected in urban and non-urban environments under varying lighting
conditions, ensuring comprehensive evaluation. Experimental results show that
TRIDENT achieves 98.8 percent accuracy in real-world recordings and 83.26
percent in a more complex setting (augmented data), outperforming unimodal and
dual-modal baselines. Moreover, TRIDENT operates in real-time, detecting drones
in just 6.09 ms while consuming only 75.27 mJ per detection, making it highly
efficient for resource-constrained devices. The dataset and code have been
released to ensure reproducibility (https://github.com/TRIDENT-2025/TRIDENT).
|
2504.06422 | Adam McArthur | Adam McArthur, Stephanie Wichuk, Stephen Burnside, Andrew Kirby,
Alexander Scammon, Damian Sol, Abhilash Hareendranathan, Jacob L. Jaremko | Retuve: Automated Multi-Modality Analysis of Hip Dysplasia with Open
Source AI | 12 pages, 8 figures, submitted to Software Impacts | null | null | null | eess.IV cs.CV | http://creativecommons.org/licenses/by/4.0/ | Developmental dysplasia of the hip (DDH) poses significant diagnostic
challenges, hindering timely intervention. Current screening methodologies lack
standardization, and AI-driven studies suffer from reproducibility issues due
to limited data and code availability. To address these limitations, we
introduce Retuve, an open-source framework for multi-modality DDH analysis,
encompassing both ultrasound (US) and X-ray imaging. Retuve provides a complete
and reproducible workflow, offering open datasets comprising expert-annotated
US and X-ray images, pre-trained models with training code and weights, and a
user-friendly Python Application Programming Interface (API). The framework
integrates segmentation and landmark detection models, enabling automated
measurement of key diagnostic parameters such as the alpha angle and acetabular
index. By adhering to open-source principles, Retuve promotes transparency,
collaboration, and accessibility in DDH research. This initiative has the
potential to democratize DDH screening, facilitate early diagnosis, and
ultimately improve patient outcomes by enabling widespread screening and early
intervention. The GitHub repository/code can be found here:
https://github.com/radoss-org/retuve
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 20:41:21 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"McArthur",
"Adam",
""
],
[
"Wichuk",
"Stephanie",
""
],
[
"Burnside",
"Stephen",
""
],
[
"Kirby",
"Andrew",
""
],
[
"Scammon",
"Alexander",
""
],
[
"Sol",
"Damian",
""
],
[
"Hareendranathan",
"Abhilash",
""
],
[
"Jaremko",
"Jacob L.",
""
]
] | TITLE: Retuve: Automated Multi-Modality Analysis of Hip Dysplasia with Open
Source AI
ABSTRACT: Developmental dysplasia of the hip (DDH) poses significant diagnostic
challenges, hindering timely intervention. Current screening methodologies lack
standardization, and AI-driven studies suffer from reproducibility issues due
to limited data and code availability. To address these limitations, we
introduce Retuve, an open-source framework for multi-modality DDH analysis,
encompassing both ultrasound (US) and X-ray imaging. Retuve provides a complete
and reproducible workflow, offering open datasets comprising expert-annotated
US and X-ray images, pre-trained models with training code and weights, and a
user-friendly Python Application Programming Interface (API). The framework
integrates segmentation and landmark detection models, enabling automated
measurement of key diagnostic parameters such as the alpha angle and acetabular
index. By adhering to open-source principles, Retuve promotes transparency,
collaboration, and accessibility in DDH research. This initiative has the
potential to democratize DDH screening, facilitate early diagnosis, and
ultimately improve patient outcomes by enabling widespread screening and early
intervention. The GitHub repository/code can be found here:
https://github.com/radoss-org/retuve
|
2504.06432 | Sibo Dong | Rupayan Mallick, Sibo Dong, Nataniel Ruiz, Sarah Adel Bargal | D-Feat Occlusions: Diffusion Features for Robustness to Partial Visual
Occlusions in Object Recognition | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Applications of diffusion models for visual tasks have been quite noteworthy.
This paper targets making classification models more robust to occlusions for
the task of object recognition by proposing a pipeline that utilizes a frozen
diffusion model. Diffusion features have demonstrated success in image
generation and image completion while understanding image context. Occlusion
can be posed as an image completion problem by deeming the pixels of the
occluder to be `missing.' We hypothesize that such features can help
hallucinate object visual features behind occluding objects, and hence we
propose using them to enable models to become more occlusion robust. We design
experiments to include input-based augmentations as well as feature-based
augmentations. Input-based augmentations involve finetuning on images where the
occluder pixels are inpainted, and feature-based augmentations involve
augmenting classification features with intermediate diffusion features. We
demonstrate that our proposed use of diffusion-based features results in models
that are more robust to partial object occlusions for both Transformers and
ConvNets on ImageNet with simulated occlusions. We also propose a dataset that
encompasses real-world occlusions and demonstrate that our method is more
robust to partial object occlusions.
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 21:05:29 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Mallick",
"Rupayan",
""
],
[
"Dong",
"Sibo",
""
],
[
"Ruiz",
"Nataniel",
""
],
[
"Bargal",
"Sarah Adel",
""
]
] | TITLE: D-Feat Occlusions: Diffusion Features for Robustness to Partial Visual
Occlusions in Object Recognition
ABSTRACT: Applications of diffusion models for visual tasks have been quite noteworthy.
This paper targets making classification models more robust to occlusions for
the task of object recognition by proposing a pipeline that utilizes a frozen
diffusion model. Diffusion features have demonstrated success in image
generation and image completion while understanding image context. Occlusion
can be posed as an image completion problem by deeming the pixels of the
occluder to be `missing.' We hypothesize that such features can help
hallucinate object visual features behind occluding objects, and hence we
propose using them to enable models to become more occlusion robust. We design
experiments to include input-based augmentations as well as feature-based
augmentations. Input-based augmentations involve finetuning on images where the
occluder pixels are inpainted, and feature-based augmentations involve
augmenting classification features with intermediate diffusion features. We
demonstrate that our proposed use of diffusion-based features results in models
that are more robust to partial object occlusions for both Transformers and
ConvNets on ImageNet with simulated occlusions. We also propose a dataset that
encompasses real-world occlusions and demonstrate that our method is more
robust to partial object occlusions.
|
2504.06460 | Hao Yan | Sai Adith Senthil Kumar, Hao Yan, Saipavan Perepa, Murong Yue, Ziyu
Yao | Can LLMs Simulate Personas with Reversed Performance? A Benchmark for
Counterfactual Instruction Following | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Large Language Models (LLMs) are now increasingly widely used to simulate
personas in virtual environments, leveraging their instruction-following
capability. However, we discovered that even state-of-the-art LLMs cannot
simulate personas with reversed performance (e.g., student personas with low
proficiency in educational settings), which impairs the simulation diversity
and limits the practical applications of the simulated environments. In this
work, using mathematical reasoning as a representative scenario, we propose the
first benchmark dataset for evaluating LLMs on simulating personas with
reversed performance, a capability that we dub "counterfactual instruction
following". We evaluate both open-weight and closed-source LLMs on this task
and find that LLMs, including the OpenAI o1 reasoning model, all struggle to
follow counterfactual instructions for simulating reversedly performing
personas. Intersectionally simulating both the performance level and the race
population of a persona worsens the effect even further. These results
highlight the challenges of counterfactual instruction following and the need
for further research.
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 22:00:32 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Kumar",
"Sai Adith Senthil",
""
],
[
"Yan",
"Hao",
""
],
[
"Perepa",
"Saipavan",
""
],
[
"Yue",
"Murong",
""
],
[
"Yao",
"Ziyu",
""
]
] | TITLE: Can LLMs Simulate Personas with Reversed Performance? A Benchmark for
Counterfactual Instruction Following
ABSTRACT: Large Language Models (LLMs) are now increasingly widely used to simulate
personas in virtual environments, leveraging their instruction-following
capability. However, we discovered that even state-of-the-art LLMs cannot
simulate personas with reversed performance (e.g., student personas with low
proficiency in educational settings), which impairs the simulation diversity
and limits the practical applications of the simulated environments. In this
work, using mathematical reasoning as a representative scenario, we propose the
first benchmark dataset for evaluating LLMs on simulating personas with
reversed performance, a capability that we dub "counterfactual instruction
following". We evaluate both open-weight and closed-source LLMs on this task
and find that LLMs, including the OpenAI o1 reasoning model, all struggle to
follow counterfactual instructions for simulating reversedly performing
personas. Intersectionally simulating both the performance level and the race
population of a persona worsens the effect even further. These results
highlight the challenges of counterfactual instruction following and the need
for further research.
|
2504.06492 | Mingchen Li | Mingchen Li, Di Zhuang, Keyu Chen, Dumindu Samaraweera, and Morris
Chang | Exploiting Meta-Learning-based Poisoning Attacks for Graph Link
Prediction | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Link prediction in graph data utilizes various algorithms and machine
learning/deep learning models to predict potential relationships between graph
nodes. This technique has found widespread use in numerous real-world
applications, including recommendation systems, community networks, and
biological structures. However, recent research has highlighted the
vulnerability of link prediction models to adversarial attacks, such as
poisoning and evasion attacks. Addressing the vulnerability of these models is
crucial to ensure stable and robust performance in link prediction
applications. While many works have focused on enhancing the robustness of the
Graph Convolution Network (GCN) model, the Variational Graph Auto-Encoder
(VGAE), a sophisticated model for link prediction, has not been thoroughly
investigated in the context of graph adversarial attacks. To bridge this gap,
this article proposes an unweighted graph poisoning attack approach using
meta-learning techniques to undermine VGAE's link prediction performance. We
conducted comprehensive experiments on diverse datasets to evaluate the
proposed method and its parameters, comparing it with existing approaches in
similar settings. Our results demonstrate that our approach significantly
diminishes link prediction performance and outperforms other state-of-the-art
methods.
| [
{
"version": "v1",
"created": "Tue, 8 Apr 2025 23:36:29 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Li",
"Mingchen",
""
],
[
"Zhuang",
"Di",
""
],
[
"Chen",
"Keyu",
""
],
[
"Samaraweera",
"Dumindu",
""
],
[
"Chang",
"Morris",
""
]
] | TITLE: Exploiting Meta-Learning-based Poisoning Attacks for Graph Link
Prediction
ABSTRACT: Link prediction in graph data utilizes various algorithms and machine
learning/deep learning models to predict potential relationships between graph
nodes. This technique has found widespread use in numerous real-world
applications, including recommendation systems, community networks, and
biological structures. However, recent research has highlighted the
vulnerability of link prediction models to adversarial attacks, such as
poisoning and evasion attacks. Addressing the vulnerability of these models is
crucial to ensure stable and robust performance in link prediction
applications. While many works have focused on enhancing the robustness of the
Graph Convolution Network (GCN) model, the Variational Graph Auto-Encoder
(VGAE), a sophisticated model for link prediction, has not been thoroughly
investigated in the context of graph adversarial attacks. To bridge this gap,
this article proposes an unweighted graph poisoning attack approach using
meta-learning techniques to undermine VGAE's link prediction performance. We
conducted comprehensive experiments on diverse datasets to evaluate the
proposed method and its parameters, comparing it with existing approaches in
similar settings. Our results demonstrate that our approach significantly
diminishes link prediction performance and outperforms other state-of-the-art
methods.
|
2504.06497 | Minati Rath | Minati Rath, Hema Date | Continuous-Variable Quantum Encoding Techniques: A Comparative Study of
Embedding Techniques and Their Impact on Machine Learning Performance | null | null | null | null | quant-ph cs.AI | http://creativecommons.org/licenses/by/4.0/ | This study explores the intersection of continuous-variable quantum computing
(CVQC) and classical machine learning, focusing on CVQC data encoding
techniques, including Displacement encoding and squeezing encoding, alongside
Instantaneous Quantum Polynomial (IQP) encoding from discrete quantum
computing. We perform an extensive empirical analysis to assess the impact of
these encoding methods on classical machine learning models, such as Logistic
Regression, Support Vector Machines, K-Nearest Neighbors, and ensemble methods
like Random Forest and LightGBM. Our findings indicate that CVQC-based encoding
methods significantly enhance feature expressivity, resulting in improved
classification accuracy and F1 scores, especially in high-dimensional and
complex datasets. However, these improvements come with varying computational
costs, which depend on the complexity of the encoding and the architecture of
the machine learning models. Additionally, we examine the trade-off between
quantum expressibility and classical learnability, offering valuable insights
into the practical feasibility of incorporating these quantum encodings into
real-world applications. This study contributes to the growing body of research
on quantum-classical hybrid learning, emphasizing the role of CVQC in advancing
quantum data representation and its integration into classical machine learning
workflows.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 00:00:45 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Rath",
"Minati",
""
],
[
"Date",
"Hema",
""
]
] | TITLE: Continuous-Variable Quantum Encoding Techniques: A Comparative Study of
Embedding Techniques and Their Impact on Machine Learning Performance
ABSTRACT: This study explores the intersection of continuous-variable quantum computing
(CVQC) and classical machine learning, focusing on CVQC data encoding
techniques, including Displacement encoding and squeezing encoding, alongside
Instantaneous Quantum Polynomial (IQP) encoding from discrete quantum
computing. We perform an extensive empirical analysis to assess the impact of
these encoding methods on classical machine learning models, such as Logistic
Regression, Support Vector Machines, K-Nearest Neighbors, and ensemble methods
like Random Forest and LightGBM. Our findings indicate that CVQC-based encoding
methods significantly enhance feature expressivity, resulting in improved
classification accuracy and F1 scores, especially in high-dimensional and
complex datasets. However, these improvements come with varying computational
costs, which depend on the complexity of the encoding and the architecture of
the machine learning models. Additionally, we examine the trade-off between
quantum expressibility and classical learnability, offering valuable insights
into the practical feasibility of incorporating these quantum encodings into
real-world applications. This study contributes to the growing body of research
on quantum-classical hybrid learning, emphasizing the role of CVQC in advancing
quantum data representation and its integration into classical machine learning
workflows.
|
2504.06504 | Xiaohang Yang | Xiaohang Yang, Qing Wang, Jiahao Yang, Gregory Slabaugh, Shanxin Yuan | STaR: Seamless Spatial-Temporal Aware Motion Retargeting with
Penetration and Consistency Constraints | 12 pages, 9 figures; | null | null | null | cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | Motion retargeting seeks to faithfully replicate the spatio-temporal motion
characteristics of a source character onto a target character with a different
body shape. Apart from motion semantics preservation, ensuring geometric
plausibility and maintaining temporal consistency are also crucial for
effective motion retargeting. However, many existing methods prioritize either
geometric plausibility or temporal consistency. Neglecting geometric
plausibility results in interpenetration while neglecting temporal consistency
leads to motion jitter. In this paper, we propose a novel sequence-to-sequence
model for seamless Spatial-Temporal aware motion Retargeting (STaR), with
penetration and consistency constraints. STaR consists of two modules: (1) a
spatial module that incorporates dense shape representation and a novel limb
penetration constraint to ensure geometric plausibility while preserving motion
semantics, and (2) a temporal module that utilizes a temporal transformer and a
novel temporal consistency constraint to predict the entire motion sequence at
once while enforcing multi-level trajectory smoothness. The seamless
combination of the two modules helps us achieve a good balance between the
semantic, geometric, and temporal targets. Extensive experiments on the Mixamo
and ScanRet datasets demonstrate that our method produces plausible and
coherent motions while significantly reducing interpenetration rates compared
with other approaches.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 00:37:08 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Yang",
"Xiaohang",
""
],
[
"Wang",
"Qing",
""
],
[
"Yang",
"Jiahao",
""
],
[
"Slabaugh",
"Gregory",
""
],
[
"Yuan",
"Shanxin",
""
]
] | TITLE: STaR: Seamless Spatial-Temporal Aware Motion Retargeting with
Penetration and Consistency Constraints
ABSTRACT: Motion retargeting seeks to faithfully replicate the spatio-temporal motion
characteristics of a source character onto a target character with a different
body shape. Apart from motion semantics preservation, ensuring geometric
plausibility and maintaining temporal consistency are also crucial for
effective motion retargeting. However, many existing methods prioritize either
geometric plausibility or temporal consistency. Neglecting geometric
plausibility results in interpenetration while neglecting temporal consistency
leads to motion jitter. In this paper, we propose a novel sequence-to-sequence
model for seamless Spatial-Temporal aware motion Retargeting (STaR), with
penetration and consistency constraints. STaR consists of two modules: (1) a
spatial module that incorporates dense shape representation and a novel limb
penetration constraint to ensure geometric plausibility while preserving motion
semantics, and (2) a temporal module that utilizes a temporal transformer and a
novel temporal consistency constraint to predict the entire motion sequence at
once while enforcing multi-level trajectory smoothness. The seamless
combination of the two modules helps us achieve a good balance between the
semantic, geometric, and temporal targets. Extensive experiments on the Mixamo
and ScanRet datasets demonstrate that our method produces plausible and
coherent motions while significantly reducing interpenetration rates compared
with other approaches.
|
2504.06511 | Tianwu Zhou | Liu Shi, Tianwu Zhou, Wei Xu, Li Liu, Zhexin Cui, Shaoyi Liang,
Haoxing Niu, Yichong Tian, Jianwei Guo | GTS-LUM: Reshaping User Behavior Modeling with LLMs in
Telecommunications Industry | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As telecommunication service providers shifting their focus to analyzing user
behavior for package design and marketing interventions, a critical challenge
lies in developing a unified, end-to-end framework capable of modeling
long-term and periodic user behavior sequences with diverse time granularities,
multi-modal data inputs, and heterogeneous labels. This paper introduces
GTS-LUM, a novel user behavior model that redefines modeling paradigms in
telecommunication settings. GTS-LUM adopts a (multi-modal) encoder-adapter-LLM
decoder architecture, enhanced with several telecom-specific innovations.
Specifically, the model incorporates an advanced timestamp processing method to
handle varying time granularities. It also supports multi-modal data inputs --
including structured tables and behavior co-occurrence graphs -- and aligns
these with semantic information extracted by a tokenizer using a Q-former
structure. Additionally, GTS-LUM integrates a front-placed target-aware
mechanism to highlight historical behaviors most relevant to the target.
Extensive experiments on industrial dataset validate the effectiveness of this
end-to-end framework and also demonstrate that GTS-LUM outperforms LLM4Rec
approaches which are popular in recommendation systems, offering an effective
and generalizing solution for user behavior modeling in telecommunications.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 01:12:07 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Shi",
"Liu",
""
],
[
"Zhou",
"Tianwu",
""
],
[
"Xu",
"Wei",
""
],
[
"Liu",
"Li",
""
],
[
"Cui",
"Zhexin",
""
],
[
"Liang",
"Shaoyi",
""
],
[
"Niu",
"Haoxing",
""
],
[
"Tian",
"Yichong",
""
],
[
"Guo",
"Jianwei",
""
]
] | TITLE: GTS-LUM: Reshaping User Behavior Modeling with LLMs in
Telecommunications Industry
ABSTRACT: As telecommunication service providers shifting their focus to analyzing user
behavior for package design and marketing interventions, a critical challenge
lies in developing a unified, end-to-end framework capable of modeling
long-term and periodic user behavior sequences with diverse time granularities,
multi-modal data inputs, and heterogeneous labels. This paper introduces
GTS-LUM, a novel user behavior model that redefines modeling paradigms in
telecommunication settings. GTS-LUM adopts a (multi-modal) encoder-adapter-LLM
decoder architecture, enhanced with several telecom-specific innovations.
Specifically, the model incorporates an advanced timestamp processing method to
handle varying time granularities. It also supports multi-modal data inputs --
including structured tables and behavior co-occurrence graphs -- and aligns
these with semantic information extracted by a tokenizer using a Q-former
structure. Additionally, GTS-LUM integrates a front-placed target-aware
mechanism to highlight historical behaviors most relevant to the target.
Extensive experiments on industrial dataset validate the effectiveness of this
end-to-end framework and also demonstrate that GTS-LUM outperforms LLM4Rec
approaches which are popular in recommendation systems, offering an effective
and generalizing solution for user behavior modeling in telecommunications.
|
2504.06514 | Ming Li | Chenrui Fan, Ming Li, Lichao Sun, Tianyi Zhou | Missing Premise exacerbates Overthinking: Are Reasoning Models losing
Critical Thinking Skill? | null | null | null | null | cs.AI cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We find that the response length of reasoning LLMs, whether trained by
reinforcement learning or supervised learning, drastically increases for
ill-posed questions with missing premises (MiP), ending up with redundant and
ineffective thinking. This newly introduced scenario exacerbates the general
overthinking issue to a large extent, which we name as the MiP-Overthinking.
Such failures are against the ``test-time scaling law'' but have been widely
observed on multiple datasets we curated with MiP, indicating the harm of cheap
overthinking and a lack of critical thinking. Surprisingly, LLMs not
specifically trained for reasoning exhibit much better performance on the MiP
scenario, producing much shorter responses that quickly identify ill-posed
queries. This implies a critical flaw of the current training recipe for
reasoning LLMs, which does not encourage efficient thinking adequately, leading
to the abuse of thinking patterns. To further investigate the reasons behind
such failures, we conduct fine-grained analyses of the reasoning length,
overthinking patterns, and location of critical thinking on different types of
LLMs. Moreover, our extended ablation study reveals that the overthinking is
contagious through the distillation of reasoning models' responses. These
results improve the understanding of overthinking and shed novel insights into
mitigating the problem.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 01:25:27 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Fan",
"Chenrui",
""
],
[
"Li",
"Ming",
""
],
[
"Sun",
"Lichao",
""
],
[
"Zhou",
"Tianyi",
""
]
] | TITLE: Missing Premise exacerbates Overthinking: Are Reasoning Models losing
Critical Thinking Skill?
ABSTRACT: We find that the response length of reasoning LLMs, whether trained by
reinforcement learning or supervised learning, drastically increases for
ill-posed questions with missing premises (MiP), ending up with redundant and
ineffective thinking. This newly introduced scenario exacerbates the general
overthinking issue to a large extent, which we name as the MiP-Overthinking.
Such failures are against the ``test-time scaling law'' but have been widely
observed on multiple datasets we curated with MiP, indicating the harm of cheap
overthinking and a lack of critical thinking. Surprisingly, LLMs not
specifically trained for reasoning exhibit much better performance on the MiP
scenario, producing much shorter responses that quickly identify ill-posed
queries. This implies a critical flaw of the current training recipe for
reasoning LLMs, which does not encourage efficient thinking adequately, leading
to the abuse of thinking patterns. To further investigate the reasons behind
such failures, we conduct fine-grained analyses of the reasoning length,
overthinking patterns, and location of critical thinking on different types of
LLMs. Moreover, our extended ablation study reveals that the overthinking is
contagious through the distillation of reasoning models' responses. These
results improve the understanding of overthinking and shed novel insights into
mitigating the problem.
|
2504.06521 | Songze Li | Songze Li, Tonghua Su, Xu-Yao Zhang, Qixing Xu, Zhongjie Wang | DUKAE: DUal-level Knowledge Accumulation and Ensemble for Pre-Trained
Model-Based Continual Learning | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Pre-trained model-based continual learning (PTMCL) has garnered growing
attention, as it enables more rapid acquisition of new knowledge by leveraging
the extensive foundational understanding inherent in pre-trained model (PTM).
Most existing PTMCL methods use Parameter-Efficient Fine-Tuning (PEFT) to learn
new knowledge while consolidating existing memory. However, they often face
some challenges. A major challenge lies in the misalignment of classification
heads, as the classification head of each task is trained within a distinct
feature space, leading to inconsistent decision boundaries across tasks and,
consequently, increased forgetting. Another critical limitation stems from the
restricted feature-level knowledge accumulation, with feature learning
typically restricted to the initial task only, which constrains the model's
representation capabilities. To address these issues, we propose a method named
DUal-level Knowledge Accumulation and Ensemble (DUKAE) that leverages both
feature-level and decision-level knowledge accumulation by aligning
classification heads into a unified feature space through Gaussian distribution
sampling and introducing an adaptive expertise ensemble to fuse knowledge
across feature subspaces.Extensive experiments on CIFAR-100, ImageNet-R,
CUB-200, and Cars-196 datasets demonstrate the superior performance of our
approach.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 01:40:38 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Li",
"Songze",
""
],
[
"Su",
"Tonghua",
""
],
[
"Zhang",
"Xu-Yao",
""
],
[
"Xu",
"Qixing",
""
],
[
"Wang",
"Zhongjie",
""
]
] | TITLE: DUKAE: DUal-level Knowledge Accumulation and Ensemble for Pre-Trained
Model-Based Continual Learning
ABSTRACT: Pre-trained model-based continual learning (PTMCL) has garnered growing
attention, as it enables more rapid acquisition of new knowledge by leveraging
the extensive foundational understanding inherent in pre-trained model (PTM).
Most existing PTMCL methods use Parameter-Efficient Fine-Tuning (PEFT) to learn
new knowledge while consolidating existing memory. However, they often face
some challenges. A major challenge lies in the misalignment of classification
heads, as the classification head of each task is trained within a distinct
feature space, leading to inconsistent decision boundaries across tasks and,
consequently, increased forgetting. Another critical limitation stems from the
restricted feature-level knowledge accumulation, with feature learning
typically restricted to the initial task only, which constrains the model's
representation capabilities. To address these issues, we propose a method named
DUal-level Knowledge Accumulation and Ensemble (DUKAE) that leverages both
feature-level and decision-level knowledge accumulation by aligning
classification heads into a unified feature space through Gaussian distribution
sampling and introducing an adaptive expertise ensemble to fuse knowledge
across feature subspaces.Extensive experiments on CIFAR-100, ImageNet-R,
CUB-200, and Cars-196 datasets demonstrate the superior performance of our
approach.
|
2504.06527 | Xyu Liu | Xinyu Liu, Xiaoguang Lin, Xiang Liu, Yong Yang, Hongqian Wang, Qilong
Sun | TSP-OCS: A Time-Series Prediction for Optimal Camera Selection in
Multi-Viewpoint Surgical Video Analysis | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Recording the open surgery process is essential for educational and medical
evaluation purposes; however, traditional single-camera methods often face
challenges such as occlusions caused by the surgeon's head and body, as well as
limitations due to fixed camera angles, which reduce comprehensibility of the
video content. This study addresses these limitations by employing a
multi-viewpoint camera recording system, capturing the surgical procedure from
six different angles to mitigate occlusions. We propose a fully supervised
learning-based time series prediction method to choose the best shot sequences
from multiple simultaneously recorded video streams, ensuring optimal
viewpoints at each moment. Our time series prediction model forecasts future
camera selections by extracting and fusing visual and semantic features from
surgical videos using pre-trained models. These features are processed by a
temporal prediction network with TimeBlocks to capture sequential dependencies.
A linear embedding layer reduces dimensionality, and a Softmax classifier
selects the optimal camera view based on the highest probability. In our
experiments, we created five groups of open thyroidectomy videos, each with
simultaneous recordings from six different angles. The results demonstrate that
our method achieves competitive accuracy compared to traditional supervised
methods, even when predicting over longer time horizons. Furthermore, our
approach outperforms state-of-the-art time series prediction techniques on our
dataset. This manuscript makes a unique contribution by presenting an
innovative framework that advances surgical video analysis techniques, with
significant implications for improving surgical education and patient safety.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 02:07:49 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Liu",
"Xinyu",
""
],
[
"Lin",
"Xiaoguang",
""
],
[
"Liu",
"Xiang",
""
],
[
"Yang",
"Yong",
""
],
[
"Wang",
"Hongqian",
""
],
[
"Sun",
"Qilong",
""
]
] | TITLE: TSP-OCS: A Time-Series Prediction for Optimal Camera Selection in
Multi-Viewpoint Surgical Video Analysis
ABSTRACT: Recording the open surgery process is essential for educational and medical
evaluation purposes; however, traditional single-camera methods often face
challenges such as occlusions caused by the surgeon's head and body, as well as
limitations due to fixed camera angles, which reduce comprehensibility of the
video content. This study addresses these limitations by employing a
multi-viewpoint camera recording system, capturing the surgical procedure from
six different angles to mitigate occlusions. We propose a fully supervised
learning-based time series prediction method to choose the best shot sequences
from multiple simultaneously recorded video streams, ensuring optimal
viewpoints at each moment. Our time series prediction model forecasts future
camera selections by extracting and fusing visual and semantic features from
surgical videos using pre-trained models. These features are processed by a
temporal prediction network with TimeBlocks to capture sequential dependencies.
A linear embedding layer reduces dimensionality, and a Softmax classifier
selects the optimal camera view based on the highest probability. In our
experiments, we created five groups of open thyroidectomy videos, each with
simultaneous recordings from six different angles. The results demonstrate that
our method achieves competitive accuracy compared to traditional supervised
methods, even when predicting over longer time horizons. Furthermore, our
approach outperforms state-of-the-art time series prediction techniques on our
dataset. This manuscript makes a unique contribution by presenting an
innovative framework that advances surgical video analysis techniques, with
significant implications for improving surgical education and patient safety.
|
2504.06529 | Khai Phan Tran | Khai Phan Tran, Xue Li | CDER: Collaborative Evidence Retrieval for Document-level Relation
Extraction | Published at ACIIDS 2024 | null | 10.1007/978-981-97-4982-9_3 | null | cs.CL | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Document-level Relation Extraction (DocRE) involves identifying relations
between entities across multiple sentences in a document. Evidence sentences,
crucial for precise entity pair relationships identification, enhance focus on
essential text segments, improving DocRE performance. However, existing
evidence retrieval systems often overlook the collaborative nature among
semantically similar entity pairs in the same document, hindering the
effectiveness of the evidence retrieval task. To address this, we propose a
novel evidence retrieval framework, namely CDER. CDER employs an attentional
graph-based architecture to capture collaborative patterns and incorporates a
dynamic sub-structure for additional robustness in evidence retrieval.
Experimental results on the benchmark DocRE dataset show that CDER not only
excels in the evidence retrieval task but also enhances overall performance of
existing DocRE system.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 02:10:21 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Tran",
"Khai Phan",
""
],
[
"Li",
"Xue",
""
]
] | TITLE: CDER: Collaborative Evidence Retrieval for Document-level Relation
Extraction
ABSTRACT: Document-level Relation Extraction (DocRE) involves identifying relations
between entities across multiple sentences in a document. Evidence sentences,
crucial for precise entity pair relationships identification, enhance focus on
essential text segments, improving DocRE performance. However, existing
evidence retrieval systems often overlook the collaborative nature among
semantically similar entity pairs in the same document, hindering the
effectiveness of the evidence retrieval task. To address this, we propose a
novel evidence retrieval framework, namely CDER. CDER employs an attentional
graph-based architecture to capture collaborative patterns and incorporates a
dynamic sub-structure for additional robustness in evidence retrieval.
Experimental results on the benchmark DocRE dataset show that CDER not only
excels in the evidence retrieval task but also enhances overall performance of
existing DocRE system.
|
2504.06533 | Zhouyang Liu | Zhouyang Liu, Ning Liu, Yixin Chen, Jiezhong He, Dongsheng Li | Flexible Graph Similarity Computation With A Proactive Optimization
Strategy | null | null | null | null | cs.LG cs.AI cs.DS | http://creativecommons.org/licenses/by/4.0/ | Graph Edit Distance (GED) is an important similarity measure in graph
retrieval, which quantifies the minimum cost of transforming one graph into
another through edit operations, and offers flexibility by allowing
customizable operation costs. Recent learning-based approaches approximate GEDs
with the distances between representations in vector spaces. However, these
methods often struggle with varying operation costs due to neglecting the
impact of these costs on determining optimal graph mappings. Furthermore, they
rely on isolated node distances as guidance, necessitating inefficient reactive
refinements of mappings. To address these issues, we propose Graph Edit Network
(GEN), a novel learning-based approach for flexible GED computation. By
identifying the limitations of existing methods in capturing flexibility of
GED, we introduce a principled yet simple solution that incorporates the
operation costs before establishing mappings. To improve matching efficiency,
we propose a strategy that proactively optimizes guidance from a graph
perspective. This strategy initializes guidance as each node's alignment
difficulty and captures the interdependencies between matches within and across
graphs through a difficulty propagation mechanism, enabling more informed
decisions. As a result, GEN selects optimal matches in a single step,
minimizing the need for costly refinements. Results on real-world and synthetic
datasets demonstrate the effectiveness, time efficiency, and adaptability of
GEN, achieving up to 37.8\% error reduction and 72.7\% inference time reduction
compared with state-of-the-art models, while performing robustly under varying
cost settings and graph sizes.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 02:16:46 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Liu",
"Zhouyang",
""
],
[
"Liu",
"Ning",
""
],
[
"Chen",
"Yixin",
""
],
[
"He",
"Jiezhong",
""
],
[
"Li",
"Dongsheng",
""
]
] | TITLE: Flexible Graph Similarity Computation With A Proactive Optimization
Strategy
ABSTRACT: Graph Edit Distance (GED) is an important similarity measure in graph
retrieval, which quantifies the minimum cost of transforming one graph into
another through edit operations, and offers flexibility by allowing
customizable operation costs. Recent learning-based approaches approximate GEDs
with the distances between representations in vector spaces. However, these
methods often struggle with varying operation costs due to neglecting the
impact of these costs on determining optimal graph mappings. Furthermore, they
rely on isolated node distances as guidance, necessitating inefficient reactive
refinements of mappings. To address these issues, we propose Graph Edit Network
(GEN), a novel learning-based approach for flexible GED computation. By
identifying the limitations of existing methods in capturing flexibility of
GED, we introduce a principled yet simple solution that incorporates the
operation costs before establishing mappings. To improve matching efficiency,
we propose a strategy that proactively optimizes guidance from a graph
perspective. This strategy initializes guidance as each node's alignment
difficulty and captures the interdependencies between matches within and across
graphs through a difficulty propagation mechanism, enabling more informed
decisions. As a result, GEN selects optimal matches in a single step,
minimizing the need for costly refinements. Results on real-world and synthetic
datasets demonstrate the effectiveness, time efficiency, and adaptability of
GEN, achieving up to 37.8\% error reduction and 72.7\% inference time reduction
compared with state-of-the-art models, while performing robustly under varying
cost settings and graph sizes.
|
2504.06536 | Happy Buzaaba | Happy Buzaaba, Alexander Wettig, David Ifeoluwa Adelani, Christiane
Fellbaum | Lugha-Llama: Adapting Large Language Models for African Languages | null | null | null | null | cs.CL cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Large language models (LLMs) have achieved impressive results in a wide range
of natural language applications. However, they often struggle to recognize
low-resource languages, in particular African languages, which are not well
represented in large training corpora. In this paper, we consider how to adapt
LLMs to low-resource African languages. We find that combining curated data
from African languages with high-quality English educational texts results in a
training mix that substantially improves the model's performance on these
languages. On the challenging IrokoBench dataset, our models consistently
achieve the best performance amongst similarly sized baselines, particularly on
knowledge-intensive multiple-choice questions (AfriMMLU). Additionally, on the
cross-lingual question answering benchmark AfriQA, our models outperform the
base model by over 10%. To better understand the role of English data during
training, we translate a subset of 200M tokens into Swahili language and
perform an analysis which reveals that the content of these data is primarily
responsible for the strong performance. We release our models and data to
encourage future research on African languages.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 02:25:53 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Buzaaba",
"Happy",
""
],
[
"Wettig",
"Alexander",
""
],
[
"Adelani",
"David Ifeoluwa",
""
],
[
"Fellbaum",
"Christiane",
""
]
] | TITLE: Lugha-Llama: Adapting Large Language Models for African Languages
ABSTRACT: Large language models (LLMs) have achieved impressive results in a wide range
of natural language applications. However, they often struggle to recognize
low-resource languages, in particular African languages, which are not well
represented in large training corpora. In this paper, we consider how to adapt
LLMs to low-resource African languages. We find that combining curated data
from African languages with high-quality English educational texts results in a
training mix that substantially improves the model's performance on these
languages. On the challenging IrokoBench dataset, our models consistently
achieve the best performance amongst similarly sized baselines, particularly on
knowledge-intensive multiple-choice questions (AfriMMLU). Additionally, on the
cross-lingual question answering benchmark AfriQA, our models outperform the
base model by over 10%. To better understand the role of English data during
training, we translate a subset of 200M tokens into Swahili language and
perform an analysis which reveals that the content of these data is primarily
responsible for the strong performance. We release our models and data to
encourage future research on African languages.
|
2504.06543 | Wei Huang | Wei Huang, Meiyu Liang, Peining Li, Xu Hou, Yawen Li, Junping Du, Zhe
Xue, Zeli Guan | DiffusionCom: Structure-Aware Multimodal Diffusion Model for Multimodal
Knowledge Graph Completion | 11 pages, 6 figures | null | null | null | cs.IR | http://creativecommons.org/licenses/by/4.0/ | Most current MKGC approaches are predominantly based on discriminative models
that maximize conditional likelihood. These approaches struggle to efficiently
capture the complex connections in real-world knowledge graphs, thereby
limiting their overall performance. To address this issue, we propose a
structure-aware multimodal Diffusion model for multimodal knowledge graph
Completion (DiffusionCom). DiffusionCom innovatively approaches the problem
from the perspective of generative models, modeling the association between the
$(head, relation)$ pair and candidate tail entities as their joint probability
distribution $p((head, relation), (tail))$, and framing the MKGC task as a
process of gradually generating the joint probability distribution from noise.
Furthermore, to fully leverage the structural information in MKGs, we propose
Structure-MKGformer, an adaptive and structure-aware multimodal knowledge
representation learning method, as the encoder for DiffusionCom.
Structure-MKGformer captures rich structural information through a multimodal
graph attention network (MGAT) and adaptively fuses it with entity
representations, thereby enhancing the structural awareness of these
representations. This design effectively addresses the limitations of existing
MKGC methods, particularly those based on multimodal pre-trained models, in
utilizing structural information. DiffusionCom is trained using both generative
and discriminative losses for the generator, while the feature extractor is
optimized exclusively with discriminative loss. This dual approach allows
DiffusionCom to harness the strengths of both generative and discriminative
models. Extensive experiments on the FB15k-237-IMG and WN18-IMG datasets
demonstrate that DiffusionCom outperforms state-of-the-art models.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 02:50:37 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Huang",
"Wei",
""
],
[
"Liang",
"Meiyu",
""
],
[
"Li",
"Peining",
""
],
[
"Hou",
"Xu",
""
],
[
"Li",
"Yawen",
""
],
[
"Du",
"Junping",
""
],
[
"Xue",
"Zhe",
""
],
[
"Guan",
"Zeli",
""
]
] | TITLE: DiffusionCom: Structure-Aware Multimodal Diffusion Model for Multimodal
Knowledge Graph Completion
ABSTRACT: Most current MKGC approaches are predominantly based on discriminative models
that maximize conditional likelihood. These approaches struggle to efficiently
capture the complex connections in real-world knowledge graphs, thereby
limiting their overall performance. To address this issue, we propose a
structure-aware multimodal Diffusion model for multimodal knowledge graph
Completion (DiffusionCom). DiffusionCom innovatively approaches the problem
from the perspective of generative models, modeling the association between the
$(head, relation)$ pair and candidate tail entities as their joint probability
distribution $p((head, relation), (tail))$, and framing the MKGC task as a
process of gradually generating the joint probability distribution from noise.
Furthermore, to fully leverage the structural information in MKGs, we propose
Structure-MKGformer, an adaptive and structure-aware multimodal knowledge
representation learning method, as the encoder for DiffusionCom.
Structure-MKGformer captures rich structural information through a multimodal
graph attention network (MGAT) and adaptively fuses it with entity
representations, thereby enhancing the structural awareness of these
representations. This design effectively addresses the limitations of existing
MKGC methods, particularly those based on multimodal pre-trained models, in
utilizing structural information. DiffusionCom is trained using both generative
and discriminative losses for the generator, while the feature extractor is
optimized exclusively with discriminative loss. This dual approach allows
DiffusionCom to harness the strengths of both generative and discriminative
models. Extensive experiments on the FB15k-237-IMG and WN18-IMG datasets
demonstrate that DiffusionCom outperforms state-of-the-art models.
|
2504.06544 | Yue Cheng | Weiwei Xing and Yue Cheng and Hongzhu Yi and Xiaohui Gao and Xiang Wei
and Xiaoyu Guo and Yuming Zhang and Xinyu Pang | LCGC: Learning from Consistency Gradient Conflicting for
Class-Imbalanced Semi-Supervised Debiasing | This paper has been accepted by AAAI 2025 | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Classifiers often learn to be biased corresponding to the class-imbalanced
dataset, especially under the semi-supervised learning (SSL) set. While
previous work tries to appropriately re-balance the classifiers by subtracting
a class-irrelevant image's logit, but lacks a firm theoretical basis. We
theoretically analyze why exploiting a baseline image can refine pseudo-labels
and prove that the black image is the best choice. We also indicated that as
the training process deepens, the pseudo-labels before and after refinement
become closer. Based on this observation, we propose a debiasing scheme dubbed
LCGC, which Learning from Consistency Gradient Conflicting, by encouraging
biased class predictions during training. We intentionally update the
pseudo-labels whose gradient conflicts with the debiased logits, representing
the optimization direction offered by the over-imbalanced classifier
predictions. Then, we debiased the predictions by subtracting the baseline
image logits during testing. Extensive experiments demonstrate that LCGC can
significantly improve the prediction accuracy of existing CISSL models on
public benchmarks.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 02:57:53 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Xing",
"Weiwei",
""
],
[
"Cheng",
"Yue",
""
],
[
"Yi",
"Hongzhu",
""
],
[
"Gao",
"Xiaohui",
""
],
[
"Wei",
"Xiang",
""
],
[
"Guo",
"Xiaoyu",
""
],
[
"Zhang",
"Yuming",
""
],
[
"Pang",
"Xinyu",
""
]
] | TITLE: LCGC: Learning from Consistency Gradient Conflicting for
Class-Imbalanced Semi-Supervised Debiasing
ABSTRACT: Classifiers often learn to be biased corresponding to the class-imbalanced
dataset, especially under the semi-supervised learning (SSL) set. While
previous work tries to appropriately re-balance the classifiers by subtracting
a class-irrelevant image's logit, but lacks a firm theoretical basis. We
theoretically analyze why exploiting a baseline image can refine pseudo-labels
and prove that the black image is the best choice. We also indicated that as
the training process deepens, the pseudo-labels before and after refinement
become closer. Based on this observation, we propose a debiasing scheme dubbed
LCGC, which Learning from Consistency Gradient Conflicting, by encouraging
biased class predictions during training. We intentionally update the
pseudo-labels whose gradient conflicts with the debiased logits, representing
the optimization direction offered by the over-imbalanced classifier
predictions. Then, we debiased the predictions by subtracting the baseline
image logits during testing. Extensive experiments demonstrate that LCGC can
significantly improve the prediction accuracy of existing CISSL models on
public benchmarks.
|
2504.06559 | Ali Eslamian | Ali Eslamian, Alireza Afzal Aghaei and Qiang Cheng | TabKAN: Advancing Tabular Data Analysis using Kolmograv-Arnold Network | 27 pages, 12 figures, 13 tables | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Tabular data analysis presents unique challenges due to its heterogeneous
feature types, missing values, and complex interactions. While traditional
machine learning methods, such as gradient boosting, often outperform deep
learning approaches, recent advancements in neural architectures offer
promising alternatives. This paper introduces TabKAN, a novel framework that
advances tabular data modeling using Kolmogorov-Arnold Networks (KANs). Unlike
conventional deep learning models, KANs leverage learnable activation functions
on edges, enhancing both interpretability and training efficiency. Our
contributions include: (1) the introduction of modular KAN-based architectures
tailored for tabular data analysis, (2) the development of a transfer learning
framework for KAN models, enabling effective knowledge transfer between
domains, (3) the development of model-specific interpretability for tabular
data learning, reducing reliance on post hoc and model-agnostic analysis, and
(4) comprehensive evaluation of vanilla supervised learning across binary and
multi-class classification tasks. Through extensive benchmarking on diverse
public datasets, TabKAN demonstrates superior performance in supervised
learning while significantly outperforming classical and Transformer-based
models in transfer learning scenarios. Our findings highlight the advantage of
KAN-based architectures in efficiently transferring knowledge across domains,
bridging the gap between traditional machine learning and deep learning for
structured data.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 03:46:10 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Eslamian",
"Ali",
""
],
[
"Aghaei",
"Alireza Afzal",
""
],
[
"Cheng",
"Qiang",
""
]
] | TITLE: TabKAN: Advancing Tabular Data Analysis using Kolmograv-Arnold Network
ABSTRACT: Tabular data analysis presents unique challenges due to its heterogeneous
feature types, missing values, and complex interactions. While traditional
machine learning methods, such as gradient boosting, often outperform deep
learning approaches, recent advancements in neural architectures offer
promising alternatives. This paper introduces TabKAN, a novel framework that
advances tabular data modeling using Kolmogorov-Arnold Networks (KANs). Unlike
conventional deep learning models, KANs leverage learnable activation functions
on edges, enhancing both interpretability and training efficiency. Our
contributions include: (1) the introduction of modular KAN-based architectures
tailored for tabular data analysis, (2) the development of a transfer learning
framework for KAN models, enabling effective knowledge transfer between
domains, (3) the development of model-specific interpretability for tabular
data learning, reducing reliance on post hoc and model-agnostic analysis, and
(4) comprehensive evaluation of vanilla supervised learning across binary and
multi-class classification tasks. Through extensive benchmarking on diverse
public datasets, TabKAN demonstrates superior performance in supervised
learning while significantly outperforming classical and Transformer-based
models in transfer learning scenarios. Our findings highlight the advantage of
KAN-based architectures in efficiently transferring knowledge across domains,
bridging the gap between traditional machine learning and deep learning for
structured data.
|
2504.06561 | Xiaohang Jiang | Xiao-Hang Jiang, Yang Ai, Rui-Chen Zheng, Zhen-Hua Ling | A Streamable Neural Audio Codec with Residual Scalar-Vector Quantization
for Real-Time Communication | Accepted by IEEE Signal Processing Letters | null | null | null | cs.SD | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper proposes StreamCodec, a streamable neural audio codec designed for
real-time communication. StreamCodec adopts a fully causal, symmetric
encoder-decoder structure and operates in the modified discrete cosine
transform (MDCT) domain, aiming for low-latency inference and real-time
efficient generation. To improve codebook utilization efficiency and compensate
for the audio quality loss caused by structural causality, StreamCodec
introduces a novel residual scalar-vector quantizer (RSVQ). The RSVQ
sequentially connects scalar quantizers and improved vector quantizers in a
residual manner, constructing coarse audio contours and refining acoustic
details, respectively. Experimental results confirm that the proposed
StreamCodec achieves decoded audio quality comparable to advanced
non-streamable neural audio codecs. Specifically, on the 16 kHz LibriTTS
dataset, StreamCodec attains a ViSQOL score of 4.30 at 1.5 kbps. It has a fixed
latency of only 20 ms and achieves a generation speed nearly 20 times real-time
on a CPU, with a lightweight model size of just 7M parameters, making it highly
suitable for real-time communication applications.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 03:49:00 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Jiang",
"Xiao-Hang",
""
],
[
"Ai",
"Yang",
""
],
[
"Zheng",
"Rui-Chen",
""
],
[
"Ling",
"Zhen-Hua",
""
]
] | TITLE: A Streamable Neural Audio Codec with Residual Scalar-Vector Quantization
for Real-Time Communication
ABSTRACT: This paper proposes StreamCodec, a streamable neural audio codec designed for
real-time communication. StreamCodec adopts a fully causal, symmetric
encoder-decoder structure and operates in the modified discrete cosine
transform (MDCT) domain, aiming for low-latency inference and real-time
efficient generation. To improve codebook utilization efficiency and compensate
for the audio quality loss caused by structural causality, StreamCodec
introduces a novel residual scalar-vector quantizer (RSVQ). The RSVQ
sequentially connects scalar quantizers and improved vector quantizers in a
residual manner, constructing coarse audio contours and refining acoustic
details, respectively. Experimental results confirm that the proposed
StreamCodec achieves decoded audio quality comparable to advanced
non-streamable neural audio codecs. Specifically, on the 16 kHz LibriTTS
dataset, StreamCodec attains a ViSQOL score of 4.30 at 1.5 kbps. It has a fixed
latency of only 20 ms and achieves a generation speed nearly 20 times real-time
on a CPU, with a lightweight model size of just 7M parameters, making it highly
suitable for real-time communication applications.
|
2504.06578 | Rahul Singh Maharjan | Rahul Singh Maharjan, Marta Romeo, Angelo Cangelosi | Attributes-aware Visual Emotion Representation Learning | 9 pages, 3 figures | null | null | null | cs.CV cs.AI cs.MM | http://creativecommons.org/licenses/by/4.0/ | Visual emotion analysis or recognition has gained considerable attention due
to the growing interest in understanding how images can convey rich semantics
and evoke emotions in human perception. However, visual emotion analysis poses
distinctive challenges compared to traditional vision tasks, especially due to
the intricate relationship between general visual features and the different
affective states they evoke, known as the affective gap. Researchers have used
deep representation learning methods to address this challenge of extracting
generalized features from entire images. However, most existing methods
overlook the importance of specific emotional attributes such as brightness,
colorfulness, scene understanding, and facial expressions. Through this paper,
we introduce A4Net, a deep representation network to bridge the affective gap
by leveraging four key attributes: brightness (Attribute 1), colorfulness
(Attribute 2), scene context (Attribute 3), and facial expressions (Attribute
4). By fusing and jointly training all aspects of attribute recognition and
visual emotion analysis, A4Net aims to provide a better insight into emotional
content in images. Experimental results show the effectiveness of A4Net,
showcasing competitive performance compared to state-of-the-art methods across
diverse visual emotion datasets. Furthermore, visualizations of activation maps
generated by A4Net offer insights into its ability to generalize across
different visual emotion datasets.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 05:00:43 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Maharjan",
"Rahul Singh",
""
],
[
"Romeo",
"Marta",
""
],
[
"Cangelosi",
"Angelo",
""
]
] | TITLE: Attributes-aware Visual Emotion Representation Learning
ABSTRACT: Visual emotion analysis or recognition has gained considerable attention due
to the growing interest in understanding how images can convey rich semantics
and evoke emotions in human perception. However, visual emotion analysis poses
distinctive challenges compared to traditional vision tasks, especially due to
the intricate relationship between general visual features and the different
affective states they evoke, known as the affective gap. Researchers have used
deep representation learning methods to address this challenge of extracting
generalized features from entire images. However, most existing methods
overlook the importance of specific emotional attributes such as brightness,
colorfulness, scene understanding, and facial expressions. Through this paper,
we introduce A4Net, a deep representation network to bridge the affective gap
by leveraging four key attributes: brightness (Attribute 1), colorfulness
(Attribute 2), scene context (Attribute 3), and facial expressions (Attribute
4). By fusing and jointly training all aspects of attribute recognition and
visual emotion analysis, A4Net aims to provide a better insight into emotional
content in images. Experimental results show the effectiveness of A4Net,
showcasing competitive performance compared to state-of-the-art methods across
diverse visual emotion datasets. Furthermore, visualizations of activation maps
generated by A4Net offer insights into its ability to generalize across
different visual emotion datasets.
|
2504.06580 | Joochan Kim | Joochan Kim, Minjoon Jung, Byoung-Tak Zhang | Exploring Ordinal Bias in Action Recognition for Instructional Videos | Accepted to SCSL @ ICLR 2025 | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Action recognition models have achieved promising results in understanding
instructional videos. However, they often rely on dominant, dataset-specific
action sequences rather than true video comprehension, a problem that we define
as ordinal bias. To address this issue, we propose two effective video
manipulation methods: Action Masking, which masks frames of frequently
co-occurring actions, and Sequence Shuffling, which randomizes the order of
action segments. Through comprehensive experiments, we demonstrate that current
models exhibit significant performance drops when confronted with nonstandard
action sequences, underscoring their vulnerability to ordinal bias. Our
findings emphasize the importance of rethinking evaluation strategies and
developing models capable of generalizing beyond fixed action patterns in
diverse instructional videos.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 05:03:51 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Kim",
"Joochan",
""
],
[
"Jung",
"Minjoon",
""
],
[
"Zhang",
"Byoung-Tak",
""
]
] | TITLE: Exploring Ordinal Bias in Action Recognition for Instructional Videos
ABSTRACT: Action recognition models have achieved promising results in understanding
instructional videos. However, they often rely on dominant, dataset-specific
action sequences rather than true video comprehension, a problem that we define
as ordinal bias. To address this issue, we propose two effective video
manipulation methods: Action Masking, which masks frames of frequently
co-occurring actions, and Sequence Shuffling, which randomizes the order of
action segments. Through comprehensive experiments, we demonstrate that current
models exhibit significant performance drops when confronted with nonstandard
action sequences, underscoring their vulnerability to ordinal bias. Our
findings emphasize the importance of rethinking evaluation strategies and
developing models capable of generalizing beyond fixed action patterns in
diverse instructional videos.
|
2504.06584 | Junrui Zhang | Junrui Zhang, Chenjie Wang, Jie Peng, Haoyu Li, Jianmin Ji, Yu Zhang,
and Yanyong Zhang | CAFE-AD: Cross-Scenario Adaptive Feature Enhancement for Trajectory
Planning in Autonomous Driving | ICRA 2025; first two authors contributed equally | null | null | null | cs.RO cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Imitation learning based planning tasks on the nuPlan dataset have gained
great interest due to their potential to generate human-like driving behaviors.
However, open-loop training on the nuPlan dataset tends to cause causal
confusion during closed-loop testing, and the dataset also presents a long-tail
distribution of scenarios. These issues introduce challenges for imitation
learning. To tackle these problems, we introduce CAFE-AD, a Cross-Scenario
Adaptive Feature Enhancement for Trajectory Planning in Autonomous Driving
method, designed to enhance feature representation across various scenario
types. We develop an adaptive feature pruning module that ranks feature
importance to capture the most relevant information while reducing the
interference of noisy information during training. Moreover, we propose a
cross-scenario feature interpolation module that enhances scenario information
to introduce diversity, enabling the network to alleviate over-fitting in
dominant scenarios. We evaluate our method CAFE-AD on the challenging public
nuPlan Test14-Hard closed-loop simulation benchmark. The results demonstrate
that CAFE-AD outperforms state-of-the-art methods including rule-based and
hybrid planners, and exhibits the potential in mitigating the impact of
long-tail distribution within the dataset. Additionally, we further validate
its effectiveness in real-world environments. The code and models will be made
available at https://github.com/AlniyatRui/CAFE-AD.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 05:16:29 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Zhang",
"Junrui",
""
],
[
"Wang",
"Chenjie",
""
],
[
"Peng",
"Jie",
""
],
[
"Li",
"Haoyu",
""
],
[
"Ji",
"Jianmin",
""
],
[
"Zhang",
"Yu",
""
],
[
"Zhang",
"Yanyong",
""
]
] | TITLE: CAFE-AD: Cross-Scenario Adaptive Feature Enhancement for Trajectory
Planning in Autonomous Driving
ABSTRACT: Imitation learning based planning tasks on the nuPlan dataset have gained
great interest due to their potential to generate human-like driving behaviors.
However, open-loop training on the nuPlan dataset tends to cause causal
confusion during closed-loop testing, and the dataset also presents a long-tail
distribution of scenarios. These issues introduce challenges for imitation
learning. To tackle these problems, we introduce CAFE-AD, a Cross-Scenario
Adaptive Feature Enhancement for Trajectory Planning in Autonomous Driving
method, designed to enhance feature representation across various scenario
types. We develop an adaptive feature pruning module that ranks feature
importance to capture the most relevant information while reducing the
interference of noisy information during training. Moreover, we propose a
cross-scenario feature interpolation module that enhances scenario information
to introduce diversity, enabling the network to alleviate over-fitting in
dominant scenarios. We evaluate our method CAFE-AD on the challenging public
nuPlan Test14-Hard closed-loop simulation benchmark. The results demonstrate
that CAFE-AD outperforms state-of-the-art methods including rule-based and
hybrid planners, and exhibits the potential in mitigating the impact of
long-tail distribution within the dataset. Additionally, we further validate
its effectiveness in real-world environments. The code and models will be made
available at https://github.com/AlniyatRui/CAFE-AD.
|
2504.06588 | Yiheng Xie | Yiheng Xie, Lucien Werner, Kaibo Chen, Thuy-Linh Le, Christine Ortega,
Steven Low | A Digital Twin of an Electrical Distribution Grid: SoCal 28-Bus Dataset | null | null | null | null | eess.SY cs.SY | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We provide an open-access dataset of phasor & waveform measurement units
(PMUs/WMUs) of a real-world electrical distribution network. The network
consists of diverse sets of generation resources (including solar panels, fuel
cells, natural gas generators, and utility interconnections), loads (including
large-scale electric vehicle charging, data centers, central cooling, offices),
topology changes (such as line outages and load transfers), as well as a
mixture of single- and three-phase networks. We describe a densely deployed PMU
sensor network in a distribution grid, in which all buses with non-zero power
injections are measured. This approach enables a range of applications such as
state estimation, system identification, power flow optimization, and feedback
control, several of which are discussed in this paper. Additionally, we provide
a synchronized waveform dataset which allows the analysis of harmonics,
transient events, dynamic grid impedance, and stability. Data collection
started in 2023 while new data is generated continuously and made available
online. A characterization of measurement error is provided. Finally, we
provide circuit topology and parameters as a part of the dataset. Together, the
circuit and timeseries data offer an opportunity for researchers to develop and
test algorithms on a real-world system.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 05:35:07 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Xie",
"Yiheng",
""
],
[
"Werner",
"Lucien",
""
],
[
"Chen",
"Kaibo",
""
],
[
"Le",
"Thuy-Linh",
""
],
[
"Ortega",
"Christine",
""
],
[
"Low",
"Steven",
""
]
] | TITLE: A Digital Twin of an Electrical Distribution Grid: SoCal 28-Bus Dataset
ABSTRACT: We provide an open-access dataset of phasor & waveform measurement units
(PMUs/WMUs) of a real-world electrical distribution network. The network
consists of diverse sets of generation resources (including solar panels, fuel
cells, natural gas generators, and utility interconnections), loads (including
large-scale electric vehicle charging, data centers, central cooling, offices),
topology changes (such as line outages and load transfers), as well as a
mixture of single- and three-phase networks. We describe a densely deployed PMU
sensor network in a distribution grid, in which all buses with non-zero power
injections are measured. This approach enables a range of applications such as
state estimation, system identification, power flow optimization, and feedback
control, several of which are discussed in this paper. Additionally, we provide
a synchronized waveform dataset which allows the analysis of harmonics,
transient events, dynamic grid impedance, and stability. Data collection
started in 2023 while new data is generated continuously and made available
online. A characterization of measurement error is provided. Finally, we
provide circuit topology and parameters as a part of the dataset. Together, the
circuit and timeseries data offer an opportunity for researchers to develop and
test algorithms on a real-world system.
|
2504.06607 | Onkar Krishna | Onkar Krishna and Hiroki Ohashi | Visually Similar Pair Alignment for Robust Cross-Domain Object Detection | 15 pages, Journal paper submission | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Domain gaps between training data (source) and real-world environments
(target) often degrade the performance of object detection models. Most
existing methods aim to bridge this gap by aligning features across source and
target domains but often fail to account for visual differences, such as color
or orientation, in alignment pairs. This limitation leads to less effective
domain adaptation, as the model struggles to manage both domain-specific shifts
(e.g., fog) and visual variations simultaneously. In this work, we demonstrate
for the first time, using a custom-built dataset, that aligning visually
similar pairs significantly improves domain adaptation. Based on this insight,
we propose a novel memory-based system to enhance domain alignment. This system
stores precomputed features of foreground objects and background areas from the
source domain, which are periodically updated during training. By retrieving
visually similar source features for alignment with target foreground and
background features, the model effectively addresses domain-specific
differences while reducing the impact of visual variations. Extensive
experiments across diverse domain shift scenarios validate our method's
effectiveness, achieving 53.1 mAP on Foggy Cityscapes and 62.3 on Sim10k,
surpassing prior state-of-the-art methods by 1.2 and 4.1 mAP, respectively.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 06:11:11 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Krishna",
"Onkar",
""
],
[
"Ohashi",
"Hiroki",
""
]
] | TITLE: Visually Similar Pair Alignment for Robust Cross-Domain Object Detection
ABSTRACT: Domain gaps between training data (source) and real-world environments
(target) often degrade the performance of object detection models. Most
existing methods aim to bridge this gap by aligning features across source and
target domains but often fail to account for visual differences, such as color
or orientation, in alignment pairs. This limitation leads to less effective
domain adaptation, as the model struggles to manage both domain-specific shifts
(e.g., fog) and visual variations simultaneously. In this work, we demonstrate
for the first time, using a custom-built dataset, that aligning visually
similar pairs significantly improves domain adaptation. Based on this insight,
we propose a novel memory-based system to enhance domain alignment. This system
stores precomputed features of foreground objects and background areas from the
source domain, which are periodically updated during training. By retrieving
visually similar source features for alignment with target foreground and
background features, the model effectively addresses domain-specific
differences while reducing the impact of visual variations. Extensive
experiments across diverse domain shift scenarios validate our method's
effectiveness, achieving 53.1 mAP on Foggy Cityscapes and 62.3 on Sim10k,
surpassing prior state-of-the-art methods by 1.2 and 4.1 mAP, respectively.
|
2504.06608 | Jiajun Chen | Jiajun Chen, Hongpeng Yin, Yifu Yang | A Cross-Domain Few-Shot Learning Method Based on Domain Knowledge
Mapping | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In task-based few-shot learning paradigms, it is commonly assumed that
different tasks are independently and identically distributed (i.i.d.).
However, in real-world scenarios, the distribution encountered in few-shot
learning can significantly differ from the distribution of existing data. Thus,
how to effectively leverage existing data knowledge to enable models to quickly
adapt to class variations under non-i.i.d. assumptions has emerged as a key
research challenge. To address this challenge, this paper proposes a new
cross-domain few-shot learning approach based on domain knowledge mapping,
applied consistently throughout the pre-training, training, and testing phases.
In the pre-training phase, our method integrates self-supervised and supervised
losses by maximizing mutual information, thereby mitigating mode collapse.
During the training phase, the domain knowledge mapping layer collaborates with
a domain classifier to learn both domain mapping capabilities and the ability
to assess domain adaptation difficulty. Finally, this approach is applied
during the testing phase, rapidly adapting to domain variations through
meta-training tasks on support sets, consequently enhancing the model's
capability to transfer domain knowledge effectively. Experimental validation
conducted across six datasets from diverse domains demonstrates the
effectiveness of the proposed method.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 06:11:55 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Chen",
"Jiajun",
""
],
[
"Yin",
"Hongpeng",
""
],
[
"Yang",
"Yifu",
""
]
] | TITLE: A Cross-Domain Few-Shot Learning Method Based on Domain Knowledge
Mapping
ABSTRACT: In task-based few-shot learning paradigms, it is commonly assumed that
different tasks are independently and identically distributed (i.i.d.).
However, in real-world scenarios, the distribution encountered in few-shot
learning can significantly differ from the distribution of existing data. Thus,
how to effectively leverage existing data knowledge to enable models to quickly
adapt to class variations under non-i.i.d. assumptions has emerged as a key
research challenge. To address this challenge, this paper proposes a new
cross-domain few-shot learning approach based on domain knowledge mapping,
applied consistently throughout the pre-training, training, and testing phases.
In the pre-training phase, our method integrates self-supervised and supervised
losses by maximizing mutual information, thereby mitigating mode collapse.
During the training phase, the domain knowledge mapping layer collaborates with
a domain classifier to learn both domain mapping capabilities and the ability
to assess domain adaptation difficulty. Finally, this approach is applied
during the testing phase, rapidly adapting to domain variations through
meta-training tasks on support sets, consequently enhancing the model's
capability to transfer domain knowledge effectively. Experimental validation
conducted across six datasets from diverse domains demonstrates the
effectiveness of the proposed method.
|
2504.06610 | Hacer Yalim Keles | Sumeyye Meryem Tasyurek and Tugce Kiziltepe and Hacer Yalim Keles | Disentangle and Regularize: Sign Language Production with
Articulator-Based Disentanglement and Channel-Aware Regularization | 11 pages, 4 figures, 1 table | null | null | null | cs.LG cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | In this work, we propose a simple gloss-free, transformer-based sign language
production (SLP) framework that directly maps spoken-language text to sign pose
sequences. We first train a pose autoencoder that encodes sign poses into a
compact latent space using an articulator-based disentanglement strategy, where
features corresponding to the face, right hand, left hand, and body are modeled
separately to promote structured and interpretable representation learning.
Next, a non-autoregressive transformer decoder is trained to predict these
latent representations from sentence-level text embeddings. To guide this
process, we apply channel-aware regularization by aligning predicted latent
distributions with priors extracted from the ground-truth encodings using a
KL-divergence loss. The contribution of each channel to the loss is weighted
according to its associated articulator region, enabling the model to account
for the relative importance of different articulators during training. Our
approach does not rely on gloss supervision or pretrained models, and achieves
state-of-the-art results on the PHOENIX14T dataset using only a modest training
set.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 06:14:19 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Tasyurek",
"Sumeyye Meryem",
""
],
[
"Kiziltepe",
"Tugce",
""
],
[
"Keles",
"Hacer Yalim",
""
]
] | TITLE: Disentangle and Regularize: Sign Language Production with
Articulator-Based Disentanglement and Channel-Aware Regularization
ABSTRACT: In this work, we propose a simple gloss-free, transformer-based sign language
production (SLP) framework that directly maps spoken-language text to sign pose
sequences. We first train a pose autoencoder that encodes sign poses into a
compact latent space using an articulator-based disentanglement strategy, where
features corresponding to the face, right hand, left hand, and body are modeled
separately to promote structured and interpretable representation learning.
Next, a non-autoregressive transformer decoder is trained to predict these
latent representations from sentence-level text embeddings. To guide this
process, we apply channel-aware regularization by aligning predicted latent
distributions with priors extracted from the ground-truth encodings using a
KL-divergence loss. The contribution of each channel to the loss is weighted
according to its associated articulator region, enabling the model to account
for the relative importance of different articulators during training. Our
approach does not rely on gloss supervision or pretrained models, and achieves
state-of-the-art results on the PHOENIX14T dataset using only a modest training
set.
|
2504.06622 | Diksha Sharma | Diksha Sharma, Vivek Balasaheb Sabale, Thirumalai M., Atul Kumar | Quantum neural networks facilitating quantum state classification | null | null | null | null | quant-ph cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The classification of quantum states into distinct classes poses a
significant challenge. In this study, we address this problem using quantum
neural networks in combination with a problem-inspired circuit and customised
as well as predefined ans\"{a}tz. To facilitate the resource-efficient quantum
state classification, we construct the dataset of quantum states using the
proposed problem-inspired circuit. The problem-inspired circuit incorporates
two-qubit parameterised unitary gates of varying entangling power, which is
further integrated with the ans\"{a}tz, developing an entire quantum neural
network. To demonstrate the capability of the selected ans\"{a}tz, we visualise
the mitigated barren plateaus. The designed quantum neural network demonstrates
the efficiency in binary and multi-class classification tasks. This work
establishes a foundation for the classification of multi-qubit quantum states
and offers the potential for generalisation to multi-qubit pure quantum states.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 06:42:32 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Sharma",
"Diksha",
""
],
[
"Sabale",
"Vivek Balasaheb",
""
],
[
"M.",
"Thirumalai",
""
],
[
"Kumar",
"Atul",
""
]
] | TITLE: Quantum neural networks facilitating quantum state classification
ABSTRACT: The classification of quantum states into distinct classes poses a
significant challenge. In this study, we address this problem using quantum
neural networks in combination with a problem-inspired circuit and customised
as well as predefined ans\"{a}tz. To facilitate the resource-efficient quantum
state classification, we construct the dataset of quantum states using the
proposed problem-inspired circuit. The problem-inspired circuit incorporates
two-qubit parameterised unitary gates of varying entangling power, which is
further integrated with the ans\"{a}tz, developing an entire quantum neural
network. To demonstrate the capability of the selected ans\"{a}tz, we visualise
the mitigated barren plateaus. The designed quantum neural network demonstrates
the efficiency in binary and multi-class classification tasks. This work
establishes a foundation for the classification of multi-qubit quantum states
and offers the potential for generalisation to multi-qubit pure quantum states.
|
2504.06633 | Zhelin Xu | Zhelin Xu, Atsushi Matsumura | A Serendipitous Recommendation System Considering User Curiosity | 15 pages, 3 figures, accepted as a full paper at iiWAS 2024 | null | 10.1007/978-3-031-78093-6_3 | null | cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | To address the problem of narrow recommendation ranges caused by an emphasis
on prediction accuracy, serendipitous recommendations, which consider both
usefulness and unexpectedness, have attracted attention. However, realizing
serendipitous recommendations is challenging due to the varying proportions of
usefulness and unexpectedness preferred by different users, which is influenced
by their differing desires for knowledge. In this paper, we propose a method to
estimate the proportion of usefulness and unexpectedness that each user desires
based on their curiosity, and make recommendations that match this preference.
The proposed method estimates a user's curiosity by considering both their
long-term and short-term interests. Offline experiments were conducted using
the MovieLens-1M dataset to evaluate the effectiveness of the proposed method.
The experimental results demonstrate that our method achieves the same level of
performance as state-of-the-art method while successfully providing
serendipitous recommendations.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 07:15:06 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Xu",
"Zhelin",
""
],
[
"Matsumura",
"Atsushi",
""
]
] | TITLE: A Serendipitous Recommendation System Considering User Curiosity
ABSTRACT: To address the problem of narrow recommendation ranges caused by an emphasis
on prediction accuracy, serendipitous recommendations, which consider both
usefulness and unexpectedness, have attracted attention. However, realizing
serendipitous recommendations is challenging due to the varying proportions of
usefulness and unexpectedness preferred by different users, which is influenced
by their differing desires for knowledge. In this paper, we propose a method to
estimate the proportion of usefulness and unexpectedness that each user desires
based on their curiosity, and make recommendations that match this preference.
The proposed method estimates a user's curiosity by considering both their
long-term and short-term interests. Offline experiments were conducted using
the MovieLens-1M dataset to evaluate the effectiveness of the proposed method.
The experimental results demonstrate that our method achieves the same level of
performance as state-of-the-art method while successfully providing
serendipitous recommendations.
|
2504.06634 | Junyoung Kim | Junyoung Kim, Youngrok Kim, Siyeol Jung, Donghyun Min | Crafting Query-Aware Selective Attention for Single Image
Super-Resolution | 10 pages, 5 figures, 4 tables | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Single Image Super-Resolution (SISR) reconstructs high-resolution images from
low-resolution inputs, enhancing image details. While Vision Transformer
(ViT)-based models improve SISR by capturing long-range dependencies, they
suffer from quadratic computational costs or employ selective attention
mechanisms that do not explicitly focus on query-relevant regions. Despite
these advancements, prior work has overlooked how selective attention
mechanisms should be effectively designed for SISR. We propose SSCAN, which
dynamically selects the most relevant key-value windows based on query
similarity, ensuring focused feature extraction while maintaining efficiency.
In contrast to prior approaches that apply attention globally or heuristically,
our method introduces a query-aware window selection strategy that better
aligns attention computation with important image regions. By incorporating
fixed-sized windows, SSCAN reduces memory usage and enforces linear
token-to-token complexity, making it scalable for large images. Our experiments
demonstrate that SSCAN outperforms existing attention-based SISR methods,
achieving up to 0.14 dB PSNR improvement on urban datasets, guaranteeing both
computational efficiency and reconstruction quality in SISR.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 07:17:29 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Kim",
"Junyoung",
""
],
[
"Kim",
"Youngrok",
""
],
[
"Jung",
"Siyeol",
""
],
[
"Min",
"Donghyun",
""
]
] | TITLE: Crafting Query-Aware Selective Attention for Single Image
Super-Resolution
ABSTRACT: Single Image Super-Resolution (SISR) reconstructs high-resolution images from
low-resolution inputs, enhancing image details. While Vision Transformer
(ViT)-based models improve SISR by capturing long-range dependencies, they
suffer from quadratic computational costs or employ selective attention
mechanisms that do not explicitly focus on query-relevant regions. Despite
these advancements, prior work has overlooked how selective attention
mechanisms should be effectively designed for SISR. We propose SSCAN, which
dynamically selects the most relevant key-value windows based on query
similarity, ensuring focused feature extraction while maintaining efficiency.
In contrast to prior approaches that apply attention globally or heuristically,
our method introduces a query-aware window selection strategy that better
aligns attention computation with important image regions. By incorporating
fixed-sized windows, SSCAN reduces memory usage and enforces linear
token-to-token complexity, making it scalable for large images. Our experiments
demonstrate that SSCAN outperforms existing attention-based SISR methods,
achieving up to 0.14 dB PSNR improvement on urban datasets, guaranteeing both
computational efficiency and reconstruction quality in SISR.
|
2504.06637 | Chenghao Ma | Chenghao Ma, Haihong E., Junpeng Ding, Jun Zhang, Ziyan Ma, Huang
Qing, Bofei Gao, Liang Chen, Meina Song | SCI-Reason: A Dataset with Chain-of-Thought Rationales for Complex
Multimodal Reasoning in Academic Areas | Submitted to ICCV 2025. 11 pages (including references) | null | null | null | cs.MM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large Language Models (LLMs) and Large Multimodal Models (LMMs) demonstrate
impressive problem-solving skills in many tasks and domains. However, their
ability to reason with complex images in academic domains has not been
systematically investigated. To bridge this gap, we present SCI-Reason, a
dataset for complex multimodel reasoning in academic areas. SCI-Reason aims to
test and improve the reasoning ability of large multimodal models using real
complex images in academic domains. The dataset contains 12,066 images and
12,626 question-answer pairs extracted from PubMed, divided into training,
validation and test splits. Each question-answer pair also contains an accurate
and efficient inference chain as a guide to improving the inference properties
of the dataset. With SCI-Reason, we performed a comprehensive evaluation of 8
well-known models. The best performing model, Claude-3.7-Sonnet, only achieved
an accuracy of 55.19%. Error analysis shows that more than half of the model
failures are due to breakdowns in multi-step inference chains rather than
errors in primary visual feature extraction. This finding underscores the
inherent limitations in reasoning capabilities exhibited by current multimodal
models when processing complex image analysis tasks within authentic academic
contexts. Experiments on open-source models show that SCI-Reason not only
enhances reasoning ability but also demonstrates cross-domain generalization in
VQA tasks. We also explore future applications of model inference capabilities
in this domain, highlighting its potential for future research.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 07:26:24 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Ma",
"Chenghao",
""
],
[
"E.",
"Haihong",
""
],
[
"Ding",
"Junpeng",
""
],
[
"Zhang",
"Jun",
""
],
[
"Ma",
"Ziyan",
""
],
[
"Qing",
"Huang",
""
],
[
"Gao",
"Bofei",
""
],
[
"Chen",
"Liang",
""
],
[
"Song",
"Meina",
""
]
] | TITLE: SCI-Reason: A Dataset with Chain-of-Thought Rationales for Complex
Multimodal Reasoning in Academic Areas
ABSTRACT: Large Language Models (LLMs) and Large Multimodal Models (LMMs) demonstrate
impressive problem-solving skills in many tasks and domains. However, their
ability to reason with complex images in academic domains has not been
systematically investigated. To bridge this gap, we present SCI-Reason, a
dataset for complex multimodel reasoning in academic areas. SCI-Reason aims to
test and improve the reasoning ability of large multimodal models using real
complex images in academic domains. The dataset contains 12,066 images and
12,626 question-answer pairs extracted from PubMed, divided into training,
validation and test splits. Each question-answer pair also contains an accurate
and efficient inference chain as a guide to improving the inference properties
of the dataset. With SCI-Reason, we performed a comprehensive evaluation of 8
well-known models. The best performing model, Claude-3.7-Sonnet, only achieved
an accuracy of 55.19%. Error analysis shows that more than half of the model
failures are due to breakdowns in multi-step inference chains rather than
errors in primary visual feature extraction. This finding underscores the
inherent limitations in reasoning capabilities exhibited by current multimodal
models when processing complex image analysis tasks within authentic academic
contexts. Experiments on open-source models show that SCI-Reason not only
enhances reasoning ability but also demonstrates cross-domain generalization in
VQA tasks. We also explore future applications of model inference capabilities
in this domain, highlighting its potential for future research.
|
2504.06638 | Hu Cui | Hu Cui, Tessai Hayama | HGMamba: Enhancing 3D Human Pose Estimation with a HyperGCN-Mamba
Network | accepted by IJCNN2025 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | 3D human pose lifting is a promising research area that leverages estimated
and ground-truth 2D human pose data for training. While existing approaches
primarily aim to enhance the performance of estimated 2D poses, they often
struggle when applied to ground-truth 2D pose data. We observe that achieving
accurate 3D pose reconstruction from ground-truth 2D poses requires precise
modeling of local pose structures, alongside the ability to extract robust
global spatio-temporal features. To address these challenges, we propose a
novel Hyper-GCN and Shuffle Mamba (HGMamba) block, which processes input data
through two parallel streams: Hyper-GCN and Shuffle-Mamba. The Hyper-GCN stream
models the human body structure as hypergraphs with varying levels of
granularity to effectively capture local joint dependencies. Meanwhile, the
Shuffle Mamba stream leverages a state space model to perform spatio-temporal
scanning across all joints, enabling the establishment of global dependencies.
By adaptively fusing these two representations, HGMamba achieves strong global
feature modeling while excelling at local structure modeling. We stack multiple
HGMamba blocks to create three variants of our model, allowing users to select
the most suitable configuration based on the desired speed-accuracy trade-off.
Extensive evaluations on the Human3.6M and MPI-INF-3DHP benchmark datasets
demonstrate the effectiveness of our approach. HGMamba-B achieves
state-of-the-art results, with P1 errors of 38.65 mm and 14.33 mm on the
respective datasets. Code and models are available:
https://github.com/HuCui2022/HGMamba
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 07:28:19 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Cui",
"Hu",
""
],
[
"Hayama",
"Tessai",
""
]
] | TITLE: HGMamba: Enhancing 3D Human Pose Estimation with a HyperGCN-Mamba
Network
ABSTRACT: 3D human pose lifting is a promising research area that leverages estimated
and ground-truth 2D human pose data for training. While existing approaches
primarily aim to enhance the performance of estimated 2D poses, they often
struggle when applied to ground-truth 2D pose data. We observe that achieving
accurate 3D pose reconstruction from ground-truth 2D poses requires precise
modeling of local pose structures, alongside the ability to extract robust
global spatio-temporal features. To address these challenges, we propose a
novel Hyper-GCN and Shuffle Mamba (HGMamba) block, which processes input data
through two parallel streams: Hyper-GCN and Shuffle-Mamba. The Hyper-GCN stream
models the human body structure as hypergraphs with varying levels of
granularity to effectively capture local joint dependencies. Meanwhile, the
Shuffle Mamba stream leverages a state space model to perform spatio-temporal
scanning across all joints, enabling the establishment of global dependencies.
By adaptively fusing these two representations, HGMamba achieves strong global
feature modeling while excelling at local structure modeling. We stack multiple
HGMamba blocks to create three variants of our model, allowing users to select
the most suitable configuration based on the desired speed-accuracy trade-off.
Extensive evaluations on the Human3.6M and MPI-INF-3DHP benchmark datasets
demonstrate the effectiveness of our approach. HGMamba-B achieves
state-of-the-art results, with P1 errors of 38.65 mm and 14.33 mm on the
respective datasets. Code and models are available:
https://github.com/HuCui2022/HGMamba
|
2504.06639 | Suvam Singh | Suvam Singh, Zolt\'an Harman, and Christoph H. Keitel | Dielectronic recombination studies of ions relevant to kilonovae and
non-LTE plasma | null | null | null | null | astro-ph.HE physics.atom-ph | http://creativecommons.org/licenses/by/4.0/ | This study presents calculations of rate coefficients, resonance strengths,
and cross sections for the dielectronic recombination (DR) of Y^+, Sr^+, Te^2+,
and Ce^2+--low-charge ions relevant to kilonovae and non-local thermodynamic
equilibrium (non-LTE) plasmas. Using relativistic atomic structure methods, we
computed DR rate coefficients under conditions typical of these environments.
Our results highlight the critical role of low-lying DR resonances in shaping
rate coefficients at kilonova temperatures (~ 10^4 K) and regulating
charge-state distributions. Pronounced near-threshold DR resonances
significantly influence the evolving ionization states and opacity of neutron
star merger ejecta. Comparisons with previous studies emphasize the necessity
of including high-n Rydberg states for accurate DR rate coefficients,
especially for complex heavy ions with dense energy levels. Discrepancies with
existing datasets underscore the need for refined computational techniques to
minimize uncertainties. These results provide essential input for interpreting
spectroscopic observations of neutron star mergers, including James Webb Space
Telescope data. We also put forward suitable candidates for experimental
studies, recognizing the challenges involved in such measurements. The data
presented here have potential to refine models of heavy-element
nucleosynthesis, enhance plasma simulation accuracy, and improve non-LTE plasma
modeling in astrophysical and laboratory settings.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 07:30:19 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Singh",
"Suvam",
""
],
[
"Harman",
"Zoltán",
""
],
[
"Keitel",
"Christoph H.",
""
]
] | TITLE: Dielectronic recombination studies of ions relevant to kilonovae and
non-LTE plasma
ABSTRACT: This study presents calculations of rate coefficients, resonance strengths,
and cross sections for the dielectronic recombination (DR) of Y^+, Sr^+, Te^2+,
and Ce^2+--low-charge ions relevant to kilonovae and non-local thermodynamic
equilibrium (non-LTE) plasmas. Using relativistic atomic structure methods, we
computed DR rate coefficients under conditions typical of these environments.
Our results highlight the critical role of low-lying DR resonances in shaping
rate coefficients at kilonova temperatures (~ 10^4 K) and regulating
charge-state distributions. Pronounced near-threshold DR resonances
significantly influence the evolving ionization states and opacity of neutron
star merger ejecta. Comparisons with previous studies emphasize the necessity
of including high-n Rydberg states for accurate DR rate coefficients,
especially for complex heavy ions with dense energy levels. Discrepancies with
existing datasets underscore the need for refined computational techniques to
minimize uncertainties. These results provide essential input for interpreting
spectroscopic observations of neutron star mergers, including James Webb Space
Telescope data. We also put forward suitable candidates for experimental
studies, recognizing the challenges involved in such measurements. The data
presented here have potential to refine models of heavy-element
nucleosynthesis, enhance plasma simulation accuracy, and improve non-LTE plasma
modeling in astrophysical and laboratory settings.
|
2504.06649 | Songwei Zhao | Songwei Zhao, Yuan Jiang, Zijing Zhang, Yang Yu, Hechang Chen | GRAIN: Multi-Granular and Implicit Information Aggregation Graph Neural
Network for Heterophilous Graphs | Accepted by AAAI 2025 | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Graph neural networks (GNNs) have shown significant success in learning graph
representations. However, recent studies reveal that GNNs often fail to
outperform simple MLPs on heterophilous graph tasks, where connected nodes may
differ in features or labels, challenging the homophily assumption. Existing
methods addressing this issue often overlook the importance of information
granularity and rarely consider implicit relationships between distant nodes.
To overcome these limitations, we propose the Granular and Implicit Graph
Network (GRAIN), a novel GNN model specifically designed for heterophilous
graphs. GRAIN enhances node embeddings by aggregating multi-view information at
various granularity levels and incorporating implicit data from distant,
non-neighboring nodes. This approach effectively integrates local and global
information, resulting in smoother, more accurate node representations. We also
introduce an adaptive graph information aggregator that efficiently combines
multi-granularity and implicit data, significantly improving node
representation quality, as shown by experiments on 13 datasets covering varying
homophily and heterophily. GRAIN consistently outperforms 12 state-of-the-art
models, excelling on both homophilous and heterophilous graphs.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 07:36:44 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Zhao",
"Songwei",
""
],
[
"Jiang",
"Yuan",
""
],
[
"Zhang",
"Zijing",
""
],
[
"Yu",
"Yang",
""
],
[
"Chen",
"Hechang",
""
]
] | TITLE: GRAIN: Multi-Granular and Implicit Information Aggregation Graph Neural
Network for Heterophilous Graphs
ABSTRACT: Graph neural networks (GNNs) have shown significant success in learning graph
representations. However, recent studies reveal that GNNs often fail to
outperform simple MLPs on heterophilous graph tasks, where connected nodes may
differ in features or labels, challenging the homophily assumption. Existing
methods addressing this issue often overlook the importance of information
granularity and rarely consider implicit relationships between distant nodes.
To overcome these limitations, we propose the Granular and Implicit Graph
Network (GRAIN), a novel GNN model specifically designed for heterophilous
graphs. GRAIN enhances node embeddings by aggregating multi-view information at
various granularity levels and incorporating implicit data from distant,
non-neighboring nodes. This approach effectively integrates local and global
information, resulting in smoother, more accurate node representations. We also
introduce an adaptive graph information aggregator that efficiently combines
multi-granularity and implicit data, significantly improving node
representation quality, as shown by experiments on 13 datasets covering varying
homophily and heterophily. GRAIN consistently outperforms 12 state-of-the-art
models, excelling on both homophilous and heterophilous graphs.
|
2504.06658 | Xiaohua Feng | Xiaohua Feng, Yuyuan Li, Chengye Wang, Junlin Liu, Li Zhang, Chaochao
Chen | A Neuro-inspired Interpretation of Unlearning in Large Language Models
through Sample-level Unlearning Difficulty | 16 pages | null | null | null | cs.LG cs.AI cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Driven by privacy protection laws and regulations, unlearning in Large
Language Models (LLMs) is gaining increasing attention. However, current
research often neglects the interpretability of the unlearning process,
particularly concerning sample-level unlearning difficulty. Existing studies
typically assume a uniform unlearning difficulty across samples. This
simplification risks attributing the performance of unlearning algorithms to
sample selection rather than the algorithm's design, potentially steering the
development of LLM unlearning in the wrong direction. Thus, we investigate the
relationship between LLM unlearning and sample characteristics, with a focus on
unlearning difficulty. Drawing inspiration from neuroscience, we propose a
Memory Removal Difficulty ($\mathrm{MRD}$) metric to quantify sample-level
unlearning difficulty. Using $\mathrm{MRD}$, we analyze the characteristics of
hard-to-unlearn versus easy-to-unlearn samples. Furthermore, we propose an
$\mathrm{MRD}$-based weighted sampling method to optimize existing unlearning
algorithms, which prioritizes easily forgettable samples, thereby improving
unlearning efficiency and effectiveness. We validate the proposed metric and
method using public benchmarks and datasets, with results confirming its
effectiveness.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 07:48:10 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Feng",
"Xiaohua",
""
],
[
"Li",
"Yuyuan",
""
],
[
"Wang",
"Chengye",
""
],
[
"Liu",
"Junlin",
""
],
[
"Zhang",
"Li",
""
],
[
"Chen",
"Chaochao",
""
]
] | TITLE: A Neuro-inspired Interpretation of Unlearning in Large Language Models
through Sample-level Unlearning Difficulty
ABSTRACT: Driven by privacy protection laws and regulations, unlearning in Large
Language Models (LLMs) is gaining increasing attention. However, current
research often neglects the interpretability of the unlearning process,
particularly concerning sample-level unlearning difficulty. Existing studies
typically assume a uniform unlearning difficulty across samples. This
simplification risks attributing the performance of unlearning algorithms to
sample selection rather than the algorithm's design, potentially steering the
development of LLM unlearning in the wrong direction. Thus, we investigate the
relationship between LLM unlearning and sample characteristics, with a focus on
unlearning difficulty. Drawing inspiration from neuroscience, we propose a
Memory Removal Difficulty ($\mathrm{MRD}$) metric to quantify sample-level
unlearning difficulty. Using $\mathrm{MRD}$, we analyze the characteristics of
hard-to-unlearn versus easy-to-unlearn samples. Furthermore, we propose an
$\mathrm{MRD}$-based weighted sampling method to optimize existing unlearning
algorithms, which prioritizes easily forgettable samples, thereby improving
unlearning efficiency and effectiveness. We validate the proposed metric and
method using public benchmarks and datasets, with results confirming its
effectiveness.
|
2504.06659 | Xiaohua Feng | Xiaohua Feng, Yuyuan Li, Huwei Ji, Jiaming Zhang, Li Zhang, Tianyu Du,
Chaochao Chen | Bridging the Gap Between Preference Alignment and Machine Unlearning | 17 pages | null | null | null | cs.LG cs.AI cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Despite advances in Preference Alignment (PA) for Large Language Models
(LLMs), mainstream methods like Reinforcement Learning with Human Feedback
(RLHF) face notable challenges. These approaches require high-quality datasets
of positive preference examples, which are costly to obtain and computationally
intensive due to training instability, limiting their use in low-resource
scenarios. LLM unlearning technique presents a promising alternative, by
directly removing the influence of negative examples. However, current research
has primarily focused on empirical validation, lacking systematic quantitative
analysis. To bridge this gap, we propose a framework to explore the
relationship between PA and LLM unlearning. Specifically, we introduce a
bi-level optimization-based method to quantify the impact of unlearning
specific negative examples on PA performance. Our analysis reveals that not all
negative examples contribute equally to alignment improvement when unlearned,
and the effect varies significantly across examples. Building on this insight,
we pose a crucial question: how can we optimally select and weight negative
examples for unlearning to maximize PA performance? To answer this, we propose
a framework called Unlearning to Align (U2A), which leverages bi-level
optimization to efficiently select and unlearn examples for optimal PA
performance. We validate the proposed method through extensive experiments,
with results confirming its effectiveness.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 07:49:08 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Feng",
"Xiaohua",
""
],
[
"Li",
"Yuyuan",
""
],
[
"Ji",
"Huwei",
""
],
[
"Zhang",
"Jiaming",
""
],
[
"Zhang",
"Li",
""
],
[
"Du",
"Tianyu",
""
],
[
"Chen",
"Chaochao",
""
]
] | TITLE: Bridging the Gap Between Preference Alignment and Machine Unlearning
ABSTRACT: Despite advances in Preference Alignment (PA) for Large Language Models
(LLMs), mainstream methods like Reinforcement Learning with Human Feedback
(RLHF) face notable challenges. These approaches require high-quality datasets
of positive preference examples, which are costly to obtain and computationally
intensive due to training instability, limiting their use in low-resource
scenarios. LLM unlearning technique presents a promising alternative, by
directly removing the influence of negative examples. However, current research
has primarily focused on empirical validation, lacking systematic quantitative
analysis. To bridge this gap, we propose a framework to explore the
relationship between PA and LLM unlearning. Specifically, we introduce a
bi-level optimization-based method to quantify the impact of unlearning
specific negative examples on PA performance. Our analysis reveals that not all
negative examples contribute equally to alignment improvement when unlearned,
and the effect varies significantly across examples. Building on this insight,
we pose a crucial question: how can we optimally select and weight negative
examples for unlearning to maximize PA performance? To answer this, we propose
a framework called Unlearning to Align (U2A), which leverages bi-level
optimization to efficiently select and unlearn examples for optimal PA
performance. We validate the proposed method through extensive experiments,
with results confirming its effectiveness.
|
2504.06660 | Osama Ahmad | Osama Ahmad, Zubair Khalid | Robust and Noise-resilient Long-Term Prediction of Spatiotemporal Data
Using Variational Mode Graph Neural Networks with 3D Attention | Accepted in IJCNN, 2025 | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | This paper focuses on improving the robustness of spatiotemporal long-term
prediction using a variational mode graph convolutional network (VMGCN) by
introducing 3D channel attention. The deep learning network for this task
relies on historical data inputs, yet real-time data can be corrupted by sensor
noise, altering its distribution. We model this noise as independent and
identically distributed (i.i.d.) Gaussian noise and incorporate it into the
LargeST traffic volume dataset, resulting in data with both inherent and
additive noise components. Our approach involves decomposing the corrupted
signal into modes using variational mode decomposition, followed by feeding the
data into a learning pipeline for prediction. We integrate a 3D attention
mechanism encompassing spatial, temporal, and channel attention. The spatial
and temporal attention modules learn their respective correlations, while the
channel attention mechanism is used to suppress noise and highlight the
significant modes in the spatiotemporal signals. Additionally, a learnable soft
thresholding method is implemented to exclude unimportant modes from the
feature vector, and a feature reduction method based on the signal-to-noise
ratio (SNR) is applied. We compare the performance of our approach against
baseline models, demonstrating that our method achieves superior long-term
prediction accuracy, robustness to noise, and improved performance with mode
truncation compared to the baseline models. The code of the paper is available
at https://github.com/OsamaAhmad369/VMGCN.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 07:49:45 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Ahmad",
"Osama",
""
],
[
"Khalid",
"Zubair",
""
]
] | TITLE: Robust and Noise-resilient Long-Term Prediction of Spatiotemporal Data
Using Variational Mode Graph Neural Networks with 3D Attention
ABSTRACT: This paper focuses on improving the robustness of spatiotemporal long-term
prediction using a variational mode graph convolutional network (VMGCN) by
introducing 3D channel attention. The deep learning network for this task
relies on historical data inputs, yet real-time data can be corrupted by sensor
noise, altering its distribution. We model this noise as independent and
identically distributed (i.i.d.) Gaussian noise and incorporate it into the
LargeST traffic volume dataset, resulting in data with both inherent and
additive noise components. Our approach involves decomposing the corrupted
signal into modes using variational mode decomposition, followed by feeding the
data into a learning pipeline for prediction. We integrate a 3D attention
mechanism encompassing spatial, temporal, and channel attention. The spatial
and temporal attention modules learn their respective correlations, while the
channel attention mechanism is used to suppress noise and highlight the
significant modes in the spatiotemporal signals. Additionally, a learnable soft
thresholding method is implemented to exclude unimportant modes from the
feature vector, and a feature reduction method based on the signal-to-noise
ratio (SNR) is applied. We compare the performance of our approach against
baseline models, demonstrating that our method achieves superior long-term
prediction accuracy, robustness to noise, and improved performance with mode
truncation compared to the baseline models. The code of the paper is available
at https://github.com/OsamaAhmad369/VMGCN.
|
2504.06672 | Elia Peruzzo | Elia Peruzzo, Dejia Xu, Xingqian Xu, Humphrey Shi, Nicu Sebe | RAGME: Retrieval Augmented Video Generation for Enhanced Motion Realism | Code available at: https://github.com/helia95/ragme | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Video generation is experiencing rapid growth, driven by advances in
diffusion models and the development of better and larger datasets. However,
producing high-quality videos remains challenging due to the high-dimensional
data and the complexity of the task. Recent efforts have primarily focused on
enhancing visual quality and addressing temporal inconsistencies, such as
flickering. Despite progress in these areas, the generated videos often fall
short in terms of motion complexity and physical plausibility, with many
outputs either appearing static or exhibiting unrealistic motion. In this work,
we propose a framework to improve the realism of motion in generated videos,
exploring a complementary direction to much of the existing literature.
Specifically, we advocate for the incorporation of a retrieval mechanism during
the generation phase. The retrieved videos act as grounding signals, providing
the model with demonstrations of how the objects move. Our pipeline is designed
to apply to any text-to-video diffusion model, conditioning a pretrained model
on the retrieved samples with minimal fine-tuning. We demonstrate the
superiority of our approach through established metrics, recently proposed
benchmarks, and qualitative results, and we highlight additional applications
of the framework.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 08:14:05 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Peruzzo",
"Elia",
""
],
[
"Xu",
"Dejia",
""
],
[
"Xu",
"Xingqian",
""
],
[
"Shi",
"Humphrey",
""
],
[
"Sebe",
"Nicu",
""
]
] | TITLE: RAGME: Retrieval Augmented Video Generation for Enhanced Motion Realism
ABSTRACT: Video generation is experiencing rapid growth, driven by advances in
diffusion models and the development of better and larger datasets. However,
producing high-quality videos remains challenging due to the high-dimensional
data and the complexity of the task. Recent efforts have primarily focused on
enhancing visual quality and addressing temporal inconsistencies, such as
flickering. Despite progress in these areas, the generated videos often fall
short in terms of motion complexity and physical plausibility, with many
outputs either appearing static or exhibiting unrealistic motion. In this work,
we propose a framework to improve the realism of motion in generated videos,
exploring a complementary direction to much of the existing literature.
Specifically, we advocate for the incorporation of a retrieval mechanism during
the generation phase. The retrieved videos act as grounding signals, providing
the model with demonstrations of how the objects move. Our pipeline is designed
to apply to any text-to-video diffusion model, conditioning a pretrained model
on the retrieved samples with minimal fine-tuning. We demonstrate the
superiority of our approach through established metrics, recently proposed
benchmarks, and qualitative results, and we highlight additional applications
of the framework.
|
2504.06680 | Christoph Balada | Christoph Balada, Aida Romano-Martinez, Vincent ten Cate, Katharina
Geschke, Jonas Tesarz, Paul Cla{\ss}en, Alexander K. Schuster, Dativa
Tibyampansha, Karl-Patrik Kresoja, Philipp S. Wild, Sheraz Ahmed, Andreas
Dengel | Deep Learning for Cardiovascular Risk Assessment: Proxy Features from
Carotid Sonography as Predictors of Arterial Damage | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this study, hypertension is utilized as an indicator of individual
vascular damage. This damage can be identified through machine learning
techniques, providing an early risk marker for potential major cardiovascular
events and offering valuable insights into the overall arterial condition of
individual patients. To this end, the VideoMAE deep learning model, originally
developed for video classification, was adapted by finetuning for application
in the domain of ultrasound imaging. The model was trained and tested using a
dataset comprising over 31,000 carotid sonography videos sourced from the
Gutenberg Health Study (15,010 participants), one of the largest prospective
population health studies. This adaptation facilitates the classification of
individuals as hypertensive or non-hypertensive (75.7% validation accuracy),
functioning as a proxy for detecting visual arterial damage. We demonstrate
that our machine learning model effectively captures visual features that
provide valuable insights into an individual's overall cardiovascular health.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 08:38:17 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Balada",
"Christoph",
""
],
[
"Romano-Martinez",
"Aida",
""
],
[
"Cate",
"Vincent ten",
""
],
[
"Geschke",
"Katharina",
""
],
[
"Tesarz",
"Jonas",
""
],
[
"Claßen",
"Paul",
""
],
[
"Schuster",
"Alexander K.",
""
],
[
"Tibyampansha",
"Dativa",
""
],
[
"Kresoja",
"Karl-Patrik",
""
],
[
"Wild",
"Philipp S.",
""
],
[
"Ahmed",
"Sheraz",
""
],
[
"Dengel",
"Andreas",
""
]
] | TITLE: Deep Learning for Cardiovascular Risk Assessment: Proxy Features from
Carotid Sonography as Predictors of Arterial Damage
ABSTRACT: In this study, hypertension is utilized as an indicator of individual
vascular damage. This damage can be identified through machine learning
techniques, providing an early risk marker for potential major cardiovascular
events and offering valuable insights into the overall arterial condition of
individual patients. To this end, the VideoMAE deep learning model, originally
developed for video classification, was adapted by finetuning for application
in the domain of ultrasound imaging. The model was trained and tested using a
dataset comprising over 31,000 carotid sonography videos sourced from the
Gutenberg Health Study (15,010 participants), one of the largest prospective
population health studies. This adaptation facilitates the classification of
individuals as hypertensive or non-hypertensive (75.7% validation accuracy),
functioning as a proxy for detecting visual arterial damage. We demonstrate
that our machine learning model effectively captures visual features that
provide valuable insights into an individual's overall cardiovascular health.
|
2504.06699 | Sam Jacob Jacob | Sam Jacob Jacob, Markus Mrosek, Carsten Othmer, Harald K\"ostler | Benchmarking Convolutional Neural Network and Graph Neural Network based
Surrogate Models on a Real-World Car External Aerodynamics Dataset | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Aerodynamic optimization is crucial for developing eco-friendly, aerodynamic,
and stylish cars, which requires close collaboration between aerodynamicists
and stylists, a collaboration impaired by the time-consuming nature of
aerodynamic simulations. Surrogate models offer a viable solution to reduce
this overhead, but they are untested in real-world aerodynamic datasets. We
present a comparative evaluation of two surrogate modeling approaches for
predicting drag on a real-world dataset: a Convolutional Neural Network (CNN)
model that uses a signed distance field as input and a commercial tool based on
Graph Neural Networks (GNN) that directly processes a surface mesh. In contrast
to previous studies based on datasets created from parameterized geometries,
our dataset comprises 343 geometries derived from 32 baseline vehicle
geometries across five distinct car projects, reflecting the diverse, free-form
modifications encountered in the typical vehicle development process. Our
results show that the CNN-based method achieves a mean absolute error of 2.3
drag counts, while the GNN-based method achieves 3.8. Both methods achieve
approximately 77% accuracy in predicting the direction of drag change relative
to the baseline geometry. While both methods effectively capture the broader
trends between baseline groups (set of samples derived from a single baseline
geometry), they struggle to varying extents in capturing the finer
intra-baseline group variations. In summary, our findings suggest that
aerodynamicists can effectively use both methods to predict drag in under two
minutes, which is at least 600 times faster than performing a simulation.
However, there remains room for improvement in capturing the finer details of
the geometry.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 09:04:59 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Jacob",
"Sam Jacob",
""
],
[
"Mrosek",
"Markus",
""
],
[
"Othmer",
"Carsten",
""
],
[
"Köstler",
"Harald",
""
]
] | TITLE: Benchmarking Convolutional Neural Network and Graph Neural Network based
Surrogate Models on a Real-World Car External Aerodynamics Dataset
ABSTRACT: Aerodynamic optimization is crucial for developing eco-friendly, aerodynamic,
and stylish cars, which requires close collaboration between aerodynamicists
and stylists, a collaboration impaired by the time-consuming nature of
aerodynamic simulations. Surrogate models offer a viable solution to reduce
this overhead, but they are untested in real-world aerodynamic datasets. We
present a comparative evaluation of two surrogate modeling approaches for
predicting drag on a real-world dataset: a Convolutional Neural Network (CNN)
model that uses a signed distance field as input and a commercial tool based on
Graph Neural Networks (GNN) that directly processes a surface mesh. In contrast
to previous studies based on datasets created from parameterized geometries,
our dataset comprises 343 geometries derived from 32 baseline vehicle
geometries across five distinct car projects, reflecting the diverse, free-form
modifications encountered in the typical vehicle development process. Our
results show that the CNN-based method achieves a mean absolute error of 2.3
drag counts, while the GNN-based method achieves 3.8. Both methods achieve
approximately 77% accuracy in predicting the direction of drag change relative
to the baseline geometry. While both methods effectively capture the broader
trends between baseline groups (set of samples derived from a single baseline
geometry), they struggle to varying extents in capturing the finer
intra-baseline group variations. In summary, our findings suggest that
aerodynamicists can effectively use both methods to predict drag in under two
minutes, which is at least 600 times faster than performing a simulation.
However, there remains room for improvement in capturing the finer details of
the geometry.
|
2504.06714 | Jujia Zhao | Jujia Zhao, Wenjie Wang, Chen Xu, Xiuying Wang, Zhaochun Ren, Suzan
Verberne | Unifying Search and Recommendation: A Generative Paradigm Inspired by
Information Theory | null | null | null | null | cs.IR | http://creativecommons.org/licenses/by/4.0/ | Recommender systems and search engines serve as foundational elements of
online platforms, with the former delivering information proactively and the
latter enabling users to seek information actively. Unifying both tasks in a
shared model is promising since it can enhance user modeling and item
understanding. Previous approaches mainly follow a discriminative paradigm,
utilizing shared encoders to process input features and task-specific heads to
perform each task. However, this paradigm encounters two key challenges:
gradient conflict and manual design complexity. From the information theory
perspective, these challenges potentially both stem from the same issue -- low
mutual information between the input features and task-specific outputs during
the optimization process.
To tackle these issues, we propose GenSR, a novel generative paradigm for
unifying search and recommendation (S&R), which leverages task-specific prompts
to partition the model's parameter space into subspaces, thereby enhancing
mutual information. To construct effective subspaces for each task, GenSR first
prepares informative representations for each subspace and then optimizes both
subspaces in one unified model. Specifically, GenSR consists of two main
modules: (1) Dual Representation Learning, which independently models
collaborative and semantic historical information to derive expressive item
representations; and (2) S&R Task Unifying, which utilizes contrastive learning
together with instruction tuning to generate task-specific outputs effectively.
Extensive experiments on two public datasets show GenSR outperforms
state-of-the-art methods across S&R tasks. Our work introduces a new generative
paradigm compared with previous discriminative methods and establishes its
superiority from the mutual information perspective.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 09:15:37 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Zhao",
"Jujia",
""
],
[
"Wang",
"Wenjie",
""
],
[
"Xu",
"Chen",
""
],
[
"Wang",
"Xiuying",
""
],
[
"Ren",
"Zhaochun",
""
],
[
"Verberne",
"Suzan",
""
]
] | TITLE: Unifying Search and Recommendation: A Generative Paradigm Inspired by
Information Theory
ABSTRACT: Recommender systems and search engines serve as foundational elements of
online platforms, with the former delivering information proactively and the
latter enabling users to seek information actively. Unifying both tasks in a
shared model is promising since it can enhance user modeling and item
understanding. Previous approaches mainly follow a discriminative paradigm,
utilizing shared encoders to process input features and task-specific heads to
perform each task. However, this paradigm encounters two key challenges:
gradient conflict and manual design complexity. From the information theory
perspective, these challenges potentially both stem from the same issue -- low
mutual information between the input features and task-specific outputs during
the optimization process.
To tackle these issues, we propose GenSR, a novel generative paradigm for
unifying search and recommendation (S&R), which leverages task-specific prompts
to partition the model's parameter space into subspaces, thereby enhancing
mutual information. To construct effective subspaces for each task, GenSR first
prepares informative representations for each subspace and then optimizes both
subspaces in one unified model. Specifically, GenSR consists of two main
modules: (1) Dual Representation Learning, which independently models
collaborative and semantic historical information to derive expressive item
representations; and (2) S&R Task Unifying, which utilizes contrastive learning
together with instruction tuning to generate task-specific outputs effectively.
Extensive experiments on two public datasets show GenSR outperforms
state-of-the-art methods across S&R tasks. Our work introduces a new generative
paradigm compared with previous discriminative methods and establishes its
superiority from the mutual information perspective.
|
2504.06719 | Pedro Hermosilla Casajus | Pedro Hermosilla and Christian Stippel and Leon Sick | Masked Scene Modeling: Narrowing the Gap Between Supervised and
Self-Supervised Learning in 3D Scene Understanding | Accepted at CVPR 2025 | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Self-supervised learning has transformed 2D computer vision by enabling
models trained on large, unannotated datasets to provide versatile
off-the-shelf features that perform similarly to models trained with labels.
However, in 3D scene understanding, self-supervised methods are typically only
used as a weight initialization step for task-specific fine-tuning, limiting
their utility for general-purpose feature extraction. This paper addresses this
shortcoming by proposing a robust evaluation protocol specifically designed to
assess the quality of self-supervised features for 3D scene understanding. Our
protocol uses multi-resolution feature sampling of hierarchical models to
create rich point-level representations that capture the semantic capabilities
of the model and, hence, are suitable for evaluation with linear probing and
nearest-neighbor methods. Furthermore, we introduce the first self-supervised
model that performs similarly to supervised models when only off-the-shelf
features are used in a linear probing setup. In particular, our model is
trained natively in 3D with a novel self-supervised approach based on a Masked
Scene Modeling objective, which reconstructs deep features of masked patches in
a bottom-up manner and is specifically tailored to hierarchical 3D models. Our
experiments not only demonstrate that our method achieves competitive
performance to supervised models, but also surpasses existing self-supervised
approaches by a large margin. The model and training code can be found at our
Github repository (https://github.com/phermosilla/msm).
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 09:19:49 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Hermosilla",
"Pedro",
""
],
[
"Stippel",
"Christian",
""
],
[
"Sick",
"Leon",
""
]
] | TITLE: Masked Scene Modeling: Narrowing the Gap Between Supervised and
Self-Supervised Learning in 3D Scene Understanding
ABSTRACT: Self-supervised learning has transformed 2D computer vision by enabling
models trained on large, unannotated datasets to provide versatile
off-the-shelf features that perform similarly to models trained with labels.
However, in 3D scene understanding, self-supervised methods are typically only
used as a weight initialization step for task-specific fine-tuning, limiting
their utility for general-purpose feature extraction. This paper addresses this
shortcoming by proposing a robust evaluation protocol specifically designed to
assess the quality of self-supervised features for 3D scene understanding. Our
protocol uses multi-resolution feature sampling of hierarchical models to
create rich point-level representations that capture the semantic capabilities
of the model and, hence, are suitable for evaluation with linear probing and
nearest-neighbor methods. Furthermore, we introduce the first self-supervised
model that performs similarly to supervised models when only off-the-shelf
features are used in a linear probing setup. In particular, our model is
trained natively in 3D with a novel self-supervised approach based on a Masked
Scene Modeling objective, which reconstructs deep features of masked patches in
a bottom-up manner and is specifically tailored to hierarchical 3D models. Our
experiments not only demonstrate that our method achieves competitive
performance to supervised models, but also surpasses existing self-supervised
approaches by a large margin. The model and training code can be found at our
Github repository (https://github.com/phermosilla/msm).
|
2504.06722 | Katsuya Akamatsu | Katsuya O. Akamatsu, Kenji Harada, Tsuyoshi Okubo, Naoki Kawashima | Plastic tensor networks for interpretable generative modeling | 37 pages, 16 figures | null | null | null | cs.LG cond-mat.stat-mech | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A structural optimization scheme for a single-layer nonnegative adaptive
tensor tree (NATT) that models a target probability distribution is proposed.
The NATT scheme, by construction, has the advantage that it is interpretable as
a probabilistic graphical model. We consider the NATT scheme and a recently
proposed Born machine adaptive tensor tree (BMATT) optimization scheme and
demonstrate their effectiveness on a variety of generative modeling tasks where
the objective is to infer the hidden structure of a provided dataset. Our
results show that in terms of minimizing the negative log-likelihood, the
single-layer scheme has model performance comparable to the Born machine
scheme, though not better. The tasks include deducing the structure of binary
bitwise operations, learning the internal structure of random Bayesian networks
given only visible sites, and a real-world example related to hierarchical
clustering where a cladogram is constructed from mitochondrial DNA sequences.
In doing so, we also show the importance of the choice of network topology and
the versatility of a least-mutual information criterion in selecting a
candidate structure for a tensor tree, as well as discuss aspects of these
tensor tree generative models including their information content and
interpretability.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 09:23:11 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Akamatsu",
"Katsuya O.",
""
],
[
"Harada",
"Kenji",
""
],
[
"Okubo",
"Tsuyoshi",
""
],
[
"Kawashima",
"Naoki",
""
]
] | TITLE: Plastic tensor networks for interpretable generative modeling
ABSTRACT: A structural optimization scheme for a single-layer nonnegative adaptive
tensor tree (NATT) that models a target probability distribution is proposed.
The NATT scheme, by construction, has the advantage that it is interpretable as
a probabilistic graphical model. We consider the NATT scheme and a recently
proposed Born machine adaptive tensor tree (BMATT) optimization scheme and
demonstrate their effectiveness on a variety of generative modeling tasks where
the objective is to infer the hidden structure of a provided dataset. Our
results show that in terms of minimizing the negative log-likelihood, the
single-layer scheme has model performance comparable to the Born machine
scheme, though not better. The tasks include deducing the structure of binary
bitwise operations, learning the internal structure of random Bayesian networks
given only visible sites, and a real-world example related to hierarchical
clustering where a cladogram is constructed from mitochondrial DNA sequences.
In doing so, we also show the importance of the choice of network topology and
the versatility of a least-mutual information criterion in selecting a
candidate structure for a tensor tree, as well as discuss aspects of these
tensor tree generative models including their information content and
interpretability.
|
2504.06740 | Hongkuan Zhou | Ylli Sadikaj, Hongkuan Zhou, Lavdim Halilaj, Stefan Schmid, Steffen
Staab, Claudia Plant | MultiADS: Defect-aware Supervision for Multi-type Anomaly Detection and
Segmentation in Zero-Shot Learning | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Precise optical inspection in industrial applications is crucial for
minimizing scrap rates and reducing the associated costs. Besides merely
detecting if a product is anomalous or not, it is crucial to know the distinct
type of defect, such as a bent, cut, or scratch. The ability to recognize the
"exact" defect type enables automated treatments of the anomalies in modern
production lines. Current methods are limited to solely detecting whether a
product is defective or not without providing any insights on the defect type,
nevertheless detecting and identifying multiple defects. We propose MultiADS, a
zero-shot learning approach, able to perform Multi-type Anomaly Detection and
Segmentation. The architecture of MultiADS comprises CLIP and extra linear
layers to align the visual- and textual representation in a joint feature
space. To the best of our knowledge, our proposal, is the first approach to
perform a multi-type anomaly segmentation task in zero-shot learning. Contrary
to the other baselines, our approach i) generates specific anomaly masks for
each distinct defect type, ii) learns to distinguish defect types, and iii)
simultaneously identifies multiple defect types present in an anomalous
product. Additionally, our approach outperforms zero/few-shot learning SoTA
methods on image-level and pixel-level anomaly detection and segmentation tasks
on five commonly used datasets: MVTec-AD, Visa, MPDD, MAD and Real-IAD.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 09:52:04 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Sadikaj",
"Ylli",
""
],
[
"Zhou",
"Hongkuan",
""
],
[
"Halilaj",
"Lavdim",
""
],
[
"Schmid",
"Stefan",
""
],
[
"Staab",
"Steffen",
""
],
[
"Plant",
"Claudia",
""
]
] | TITLE: MultiADS: Defect-aware Supervision for Multi-type Anomaly Detection and
Segmentation in Zero-Shot Learning
ABSTRACT: Precise optical inspection in industrial applications is crucial for
minimizing scrap rates and reducing the associated costs. Besides merely
detecting if a product is anomalous or not, it is crucial to know the distinct
type of defect, such as a bent, cut, or scratch. The ability to recognize the
"exact" defect type enables automated treatments of the anomalies in modern
production lines. Current methods are limited to solely detecting whether a
product is defective or not without providing any insights on the defect type,
nevertheless detecting and identifying multiple defects. We propose MultiADS, a
zero-shot learning approach, able to perform Multi-type Anomaly Detection and
Segmentation. The architecture of MultiADS comprises CLIP and extra linear
layers to align the visual- and textual representation in a joint feature
space. To the best of our knowledge, our proposal, is the first approach to
perform a multi-type anomaly segmentation task in zero-shot learning. Contrary
to the other baselines, our approach i) generates specific anomaly masks for
each distinct defect type, ii) learns to distinguish defect types, and iii)
simultaneously identifies multiple defect types present in an anomalous
product. Additionally, our approach outperforms zero/few-shot learning SoTA
methods on image-level and pixel-level anomaly detection and segmentation tasks
on five commonly used datasets: MVTec-AD, Visa, MPDD, MAD and Real-IAD.
|
2504.06741 | Constantin Ulrich | Constantin Ulrich, Tassilo Wald, Fabian Isensee, Klaus H. Maier-Hein | Large Scale Supervised Pretraining For Traumatic Brain Injury
Segmentation | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | The segmentation of lesions in Moderate to Severe Traumatic Brain Injury
(msTBI) presents a significant challenge in neuroimaging due to the diverse
characteristics of these lesions, which vary in size, shape, and distribution
across brain regions and tissue types. This heterogeneity complicates
traditional image processing techniques, resulting in critical errors in tasks
such as image registration and brain parcellation. To address these challenges,
the AIMS-TBI Segmentation Challenge 2024 aims to advance innovative
segmentation algorithms specifically designed for T1-weighted MRI data, the
most widely utilized imaging modality in clinical practice. Our proposed
solution leverages a large-scale multi-dataset supervised pretraining approach
inspired by the MultiTalent method. We train a Resenc L network on a
comprehensive collection of datasets covering various anatomical and
pathological structures, which equips the model with a robust understanding of
brain anatomy and pathology. Following this, the model is fine-tuned on
msTBI-specific data to optimize its performance for the unique characteristics
of T1-weighted MRI scans and outperforms the baseline without pretraining up to
2 Dice points.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 09:52:45 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Ulrich",
"Constantin",
""
],
[
"Wald",
"Tassilo",
""
],
[
"Isensee",
"Fabian",
""
],
[
"Maier-Hein",
"Klaus H.",
""
]
] | TITLE: Large Scale Supervised Pretraining For Traumatic Brain Injury
Segmentation
ABSTRACT: The segmentation of lesions in Moderate to Severe Traumatic Brain Injury
(msTBI) presents a significant challenge in neuroimaging due to the diverse
characteristics of these lesions, which vary in size, shape, and distribution
across brain regions and tissue types. This heterogeneity complicates
traditional image processing techniques, resulting in critical errors in tasks
such as image registration and brain parcellation. To address these challenges,
the AIMS-TBI Segmentation Challenge 2024 aims to advance innovative
segmentation algorithms specifically designed for T1-weighted MRI data, the
most widely utilized imaging modality in clinical practice. Our proposed
solution leverages a large-scale multi-dataset supervised pretraining approach
inspired by the MultiTalent method. We train a Resenc L network on a
comprehensive collection of datasets covering various anatomical and
pathological structures, which equips the model with a robust understanding of
brain anatomy and pathology. Following this, the model is fine-tuned on
msTBI-specific data to optimize its performance for the unique characteristics
of T1-weighted MRI scans and outperforms the baseline without pretraining up to
2 Dice points.
|
2504.06766 | Yuxin Wang | Yuxin Wang, Yiran Guo, Yining Zheng, Zhangyue Yin, Shuo Chen, Jie
Yang, Jiajun Chen, Xuanjing Huang, Xipeng Qiu | FamilyTool: A Multi-hop Personalized Tool Use Benchmark | null | null | null | null | cs.AI cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The integration of tool learning with Large Language Models (LLMs) has
expanded their capabilities in handling complex tasks by leveraging external
tools. However, existing benchmarks for tool learning inadequately address
critical real-world personalized scenarios, particularly those requiring
multi-hop reasoning and inductive knowledge adaptation in dynamic environments.
To bridge this gap, we introduce FamilyTool, a novel benchmark grounded in a
family-based knowledge graph (KG) that simulates personalized, multi-hop tool
use scenarios. FamilyTool challenges LLMs with queries spanning 1 to 3
relational hops (e.g., inferring familial connections and preferences) and
incorporates an inductive KG setting where models must adapt to unseen user
preferences and relationships without re-training, a common limitation in prior
approaches that compromises generalization. We further propose KGETool: a
simple KG-augmented evaluation pipeline to systematically assess LLMs' tool use
ability in these settings. Experiments reveal significant performance gaps in
state-of-the-art LLMs, with accuracy dropping sharply as hop complexity
increases and inductive scenarios exposing severe generalization deficits.
These findings underscore the limitations of current LLMs in handling
personalized, evolving real-world contexts and highlight the urgent need for
advancements in tool-learning frameworks. FamilyTool serves as a critical
resource for evaluating and advancing LLM agents' reasoning, adaptability, and
scalability in complex, dynamic environments. Code and dataset are available at
Github.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 10:42:36 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Wang",
"Yuxin",
""
],
[
"Guo",
"Yiran",
""
],
[
"Zheng",
"Yining",
""
],
[
"Yin",
"Zhangyue",
""
],
[
"Chen",
"Shuo",
""
],
[
"Yang",
"Jie",
""
],
[
"Chen",
"Jiajun",
""
],
[
"Huang",
"Xuanjing",
""
],
[
"Qiu",
"Xipeng",
""
]
] | TITLE: FamilyTool: A Multi-hop Personalized Tool Use Benchmark
ABSTRACT: The integration of tool learning with Large Language Models (LLMs) has
expanded their capabilities in handling complex tasks by leveraging external
tools. However, existing benchmarks for tool learning inadequately address
critical real-world personalized scenarios, particularly those requiring
multi-hop reasoning and inductive knowledge adaptation in dynamic environments.
To bridge this gap, we introduce FamilyTool, a novel benchmark grounded in a
family-based knowledge graph (KG) that simulates personalized, multi-hop tool
use scenarios. FamilyTool challenges LLMs with queries spanning 1 to 3
relational hops (e.g., inferring familial connections and preferences) and
incorporates an inductive KG setting where models must adapt to unseen user
preferences and relationships without re-training, a common limitation in prior
approaches that compromises generalization. We further propose KGETool: a
simple KG-augmented evaluation pipeline to systematically assess LLMs' tool use
ability in these settings. Experiments reveal significant performance gaps in
state-of-the-art LLMs, with accuracy dropping sharply as hop complexity
increases and inductive scenarios exposing severe generalization deficits.
These findings underscore the limitations of current LLMs in handling
personalized, evolving real-world contexts and highlight the urgent need for
advancements in tool-learning frameworks. FamilyTool serves as a critical
resource for evaluating and advancing LLM agents' reasoning, adaptability, and
scalability in complex, dynamic environments. Code and dataset are available at
Github.
|
2504.06767 | Matteo Santacesaria | Paolo Angella, Luca Balbi, Fabrizio Ferrando, Paolo Traverso, Rosario
Varriale, Vito Paolo Pastore, Matteo Santacesaria | DIMA: DIffusing Motion Artifacts for unsupervised correction in brain
MRI images | 7 pages, 5 figures, 7 tables | null | null | null | eess.IV cs.CV | http://creativecommons.org/licenses/by/4.0/ | Motion artifacts remain a significant challenge in Magnetic Resonance Imaging
(MRI), compromising diagnostic quality and potentially leading to misdiagnosis
or repeated scans. Existing deep learning approaches for motion artifact
correction typically require paired motion-free and motion-affected images for
training, which are rarely available in clinical settings. To overcome this
requirement, we present DIMA (DIffusing Motion Artifacts), a novel framework
that leverages diffusion models to enable unsupervised motion artifact
correction in brain MRI. Our two-phase approach first trains a diffusion model
on unpaired motion-affected images to learn the distribution of motion
artifacts. This model then generates realistic motion artifacts on clean
images, creating paired datasets suitable for supervised training of correction
networks. Unlike existing methods, DIMA operates without requiring k-space
manipulation or detailed knowledge of MRI sequence parameters, making it
adaptable across different scanning protocols and hardware. Comprehensive
evaluations across multiple datasets and anatomical planes demonstrate that our
method achieves comparable performance to state-of-the-art supervised
approaches while offering superior generalizability to real clinical data. DIMA
represents a significant advancement in making motion artifact correction more
accessible for routine clinical use, potentially reducing the need for repeat
scans and improving diagnostic accuracy.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 10:43:38 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Angella",
"Paolo",
""
],
[
"Balbi",
"Luca",
""
],
[
"Ferrando",
"Fabrizio",
""
],
[
"Traverso",
"Paolo",
""
],
[
"Varriale",
"Rosario",
""
],
[
"Pastore",
"Vito Paolo",
""
],
[
"Santacesaria",
"Matteo",
""
]
] | TITLE: DIMA: DIffusing Motion Artifacts for unsupervised correction in brain
MRI images
ABSTRACT: Motion artifacts remain a significant challenge in Magnetic Resonance Imaging
(MRI), compromising diagnostic quality and potentially leading to misdiagnosis
or repeated scans. Existing deep learning approaches for motion artifact
correction typically require paired motion-free and motion-affected images for
training, which are rarely available in clinical settings. To overcome this
requirement, we present DIMA (DIffusing Motion Artifacts), a novel framework
that leverages diffusion models to enable unsupervised motion artifact
correction in brain MRI. Our two-phase approach first trains a diffusion model
on unpaired motion-affected images to learn the distribution of motion
artifacts. This model then generates realistic motion artifacts on clean
images, creating paired datasets suitable for supervised training of correction
networks. Unlike existing methods, DIMA operates without requiring k-space
manipulation or detailed knowledge of MRI sequence parameters, making it
adaptable across different scanning protocols and hardware. Comprehensive
evaluations across multiple datasets and anatomical planes demonstrate that our
method achieves comparable performance to state-of-the-art supervised
approaches while offering superior generalizability to real clinical data. DIMA
represents a significant advancement in making motion artifact correction more
accessible for routine clinical use, potentially reducing the need for repeat
scans and improving diagnostic accuracy.
|
2504.06780 | Yong Bai | Yong Bai, Rui Xiang, Kaiyuan Li, Yongxiang Tang, Yanhua Cheng, Xialong
Liu, Peng Jiang, Kun Gai | CHIME: A Compressive Framework for Holistic Interest Modeling | null | null | null | null | cs.IR | http://creativecommons.org/licenses/by/4.0/ | Modeling holistic user interests is important for improving recommendation
systems but is challenged by high computational cost and difficulty in handling
diverse information with full behavior context. Existing search-based methods
might lose critical signals during behavior selection. To overcome these
limitations, we propose CHIME: A Compressive Framework for Holistic Interest
Modeling. It uses adapted large language models to encode complete user
behaviors with heterogeneous inputs. We introduce multi-granular contrastive
learning objectives to capture both persistent and transient interest patterns
and apply residual vector quantization to generate compact embeddings. CHIME
demonstrates superior ranking performance across diverse datasets, establishing
a robust solution for scalable holistic interest modeling in recommendation
systems.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 11:08:49 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Bai",
"Yong",
""
],
[
"Xiang",
"Rui",
""
],
[
"Li",
"Kaiyuan",
""
],
[
"Tang",
"Yongxiang",
""
],
[
"Cheng",
"Yanhua",
""
],
[
"Liu",
"Xialong",
""
],
[
"Jiang",
"Peng",
""
],
[
"Gai",
"Kun",
""
]
] | TITLE: CHIME: A Compressive Framework for Holistic Interest Modeling
ABSTRACT: Modeling holistic user interests is important for improving recommendation
systems but is challenged by high computational cost and difficulty in handling
diverse information with full behavior context. Existing search-based methods
might lose critical signals during behavior selection. To overcome these
limitations, we propose CHIME: A Compressive Framework for Holistic Interest
Modeling. It uses adapted large language models to encode complete user
behaviors with heterogeneous inputs. We introduce multi-granular contrastive
learning objectives to capture both persistent and transient interest patterns
and apply residual vector quantization to generate compact embeddings. CHIME
demonstrates superior ranking performance across diverse datasets, establishing
a robust solution for scalable holistic interest modeling in recommendation
systems.
|
2504.06781 | Reiji Saito | Reiji Saito, Kazuhiro Hotta | Domain Generalization through Attenuation of Domain-Specific Information | Accepted by CVPR 2025 Workshops | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we propose a new evaluation metric called Domain Independence
(DI) and Attenuation of Domain-Specific Information (ADSI) which is
specifically designed for domain-generalized semantic segmentation in
automotive images. DI measures the presence of domain-specific information: a
lower DI value indicates strong domain dependence, while a higher DI value
suggests greater domain independence. This makes it roughly where
domain-specific information exists and up to which frequency range it is
present. As a result, it becomes possible to effectively suppress only the
regions in the image that contain domain-specific information, enabling feature
extraction independent of the domain. ADSI uses a Butterworth filter to remove
the low-frequency components of images that contain inherent domain-specific
information such as sensor characteristics and lighting conditions. However,
since low-frequency components also contain important information such as
color, we should not remove them completely. Thus, a scalar value (ranging from
0 to 1) is multiplied by the low-frequency components to retain essential
information. This helps the model learn more domain-independent features. In
experiments, GTA5 (synthetic dataset) was used as training images, and a
real-world dataset was used for evaluation, and the proposed method
outperformed conventional approaches. Similarly, in experiments that the
Cityscapes (real-world dataset) was used for training and various environment
datasets such as rain and nighttime were used for evaluation, the proposed
method demonstrated its robustness under nighttime conditions.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 11:10:29 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Saito",
"Reiji",
""
],
[
"Hotta",
"Kazuhiro",
""
]
] | TITLE: Domain Generalization through Attenuation of Domain-Specific Information
ABSTRACT: In this paper, we propose a new evaluation metric called Domain Independence
(DI) and Attenuation of Domain-Specific Information (ADSI) which is
specifically designed for domain-generalized semantic segmentation in
automotive images. DI measures the presence of domain-specific information: a
lower DI value indicates strong domain dependence, while a higher DI value
suggests greater domain independence. This makes it roughly where
domain-specific information exists and up to which frequency range it is
present. As a result, it becomes possible to effectively suppress only the
regions in the image that contain domain-specific information, enabling feature
extraction independent of the domain. ADSI uses a Butterworth filter to remove
the low-frequency components of images that contain inherent domain-specific
information such as sensor characteristics and lighting conditions. However,
since low-frequency components also contain important information such as
color, we should not remove them completely. Thus, a scalar value (ranging from
0 to 1) is multiplied by the low-frequency components to retain essential
information. This helps the model learn more domain-independent features. In
experiments, GTA5 (synthetic dataset) was used as training images, and a
real-world dataset was used for evaluation, and the proposed method
outperformed conventional approaches. Similarly, in experiments that the
Cityscapes (real-world dataset) was used for training and various environment
datasets such as rain and nighttime were used for evaluation, the proposed
method demonstrated its robustness under nighttime conditions.
|
2504.06785 | Andrea Visentin Dr | Shuoshuo Xu, Kai Zhao, James Loney, Zili Li, Andrea Visentin | Zero-Shot Image-Based Large Language Model Approach to Road Pavement
Monitoring | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Effective and rapid evaluation of pavement surface condition is critical for
prioritizing maintenance, ensuring transportation safety, and minimizing
vehicle wear and tear. While conventional manual inspections suffer from
subjectivity, existing machine learning-based methods are constrained by their
reliance on large and high-quality labeled datasets, which require significant
resources and limit adaptability across varied road conditions. The
revolutionary advancements in Large Language Models (LLMs) present significant
potential for overcoming these challenges. In this study, we propose an
innovative automated zero-shot learning approach that leverages the image
recognition and natural language understanding capabilities of LLMs to assess
road conditions effectively. Multiple LLM-based assessment models were
developed, employing prompt engineering strategies aligned with the Pavement
Surface Condition Index (PSCI) standards. These models' accuracy and
reliability were evaluated against official PSCI results, with an optimized
model ultimately selected. Extensive tests benchmarked the optimized model
against evaluations from various levels experts using Google Street View road
images. The results reveal that the LLM-based approach can effectively assess
road conditions, with the optimized model -employing comprehensive and
structured prompt engineering strategies -outperforming simpler configurations
by achieving high accuracy and consistency, even surpassing expert evaluations.
Moreover, successfully applying the optimized model to Google Street View
images demonstrates its potential for future city-scale deployments. These
findings highlight the transformative potential of LLMs in automating road
damage evaluations and underscore the pivotal role of detailed prompt
engineering in achieving reliable assessments.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 11:19:17 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Xu",
"Shuoshuo",
""
],
[
"Zhao",
"Kai",
""
],
[
"Loney",
"James",
""
],
[
"Li",
"Zili",
""
],
[
"Visentin",
"Andrea",
""
]
] | TITLE: Zero-Shot Image-Based Large Language Model Approach to Road Pavement
Monitoring
ABSTRACT: Effective and rapid evaluation of pavement surface condition is critical for
prioritizing maintenance, ensuring transportation safety, and minimizing
vehicle wear and tear. While conventional manual inspections suffer from
subjectivity, existing machine learning-based methods are constrained by their
reliance on large and high-quality labeled datasets, which require significant
resources and limit adaptability across varied road conditions. The
revolutionary advancements in Large Language Models (LLMs) present significant
potential for overcoming these challenges. In this study, we propose an
innovative automated zero-shot learning approach that leverages the image
recognition and natural language understanding capabilities of LLMs to assess
road conditions effectively. Multiple LLM-based assessment models were
developed, employing prompt engineering strategies aligned with the Pavement
Surface Condition Index (PSCI) standards. These models' accuracy and
reliability were evaluated against official PSCI results, with an optimized
model ultimately selected. Extensive tests benchmarked the optimized model
against evaluations from various levels experts using Google Street View road
images. The results reveal that the LLM-based approach can effectively assess
road conditions, with the optimized model -employing comprehensive and
structured prompt engineering strategies -outperforming simpler configurations
by achieving high accuracy and consistency, even surpassing expert evaluations.
Moreover, successfully applying the optimized model to Google Street View
images demonstrates its potential for future city-scale deployments. These
findings highlight the transformative potential of LLMs in automating road
damage evaluations and underscore the pivotal role of detailed prompt
engineering in achieving reliable assessments.
|
2504.06805 | Nicola Novello | Nicola Novello and Andrea M. Tonello | Robust Classification with Noisy Labels Based on Posterior Maximization | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Designing objective functions robust to label noise is crucial for real-world
classification algorithms. In this paper, we investigate the robustness to
label noise of an $f$-divergence-based class of objective functions recently
proposed for supervised classification, herein referred to as $f$-PML. We show
that, in the presence of label noise, any of the $f$-PML objective functions
can be corrected to obtain a neural network that is equal to the one learned
with the clean dataset. Additionally, we propose an alternative and novel
correction approach that, during the test phase, refines the posterior
estimated by the neural network trained in the presence of label noise. Then,
we demonstrate that, even if the considered $f$-PML objective functions are not
symmetric, they are robust to symmetric label noise for any choice of
$f$-divergence, without the need for any correction approach. This allows us to
prove that the cross-entropy, which belongs to the $f$-PML class, is robust to
symmetric label noise. Finally, we show that such a class of objective
functions can be used together with refined training strategies, achieving
competitive performance against state-of-the-art techniques of classification
with label noise.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 11:52:51 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Novello",
"Nicola",
""
],
[
"Tonello",
"Andrea M.",
""
]
] | TITLE: Robust Classification with Noisy Labels Based on Posterior Maximization
ABSTRACT: Designing objective functions robust to label noise is crucial for real-world
classification algorithms. In this paper, we investigate the robustness to
label noise of an $f$-divergence-based class of objective functions recently
proposed for supervised classification, herein referred to as $f$-PML. We show
that, in the presence of label noise, any of the $f$-PML objective functions
can be corrected to obtain a neural network that is equal to the one learned
with the clean dataset. Additionally, we propose an alternative and novel
correction approach that, during the test phase, refines the posterior
estimated by the neural network trained in the presence of label noise. Then,
we demonstrate that, even if the considered $f$-PML objective functions are not
symmetric, they are robust to symmetric label noise for any choice of
$f$-divergence, without the need for any correction approach. This allows us to
prove that the cross-entropy, which belongs to the $f$-PML class, is robust to
symmetric label noise. Finally, we show that such a class of objective
functions can be used together with refined training strategies, achieving
competitive performance against state-of-the-art techniques of classification
with label noise.
|
2504.06811 | Bhavesh Gyanchandani | Abhinav Roy, Bhavesh Gyanchandani, and Aditya Oza | Hybrid CNN with Chebyshev Polynomial Expansion for Medical Image
Analysis | null | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Lung cancer remains one of the leading causes of cancer-related mortality
worldwide, with early and accurate diagnosis playing a pivotal role in
improving patient outcomes. Automated detection of pulmonary nodules in
computed tomography (CT) scans is a challenging task due to variability in
nodule size, shape, texture, and location. Traditional Convolutional Neural
Networks (CNNs) have shown considerable promise in medical image analysis;
however, their limited ability to capture fine-grained spatial-spectral
variations restricts their performance in complex diagnostic scenarios. In this
study, we propose a novel hybrid deep learning architecture that incorporates
Chebyshev polynomial expansions into CNN layers to enhance expressive power and
improve the representation of underlying anatomical structures. The proposed
Chebyshev-CNN leverages the orthogonality and recursive properties of Chebyshev
polynomials to extract high-frequency features and approximate complex
nonlinear functions with greater fidelity. The model is trained and evaluated
on benchmark lung cancer imaging datasets, including LUNA16 and LIDC-IDRI,
achieving superior performance in classifying pulmonary nodules as benign or
malignant. Quantitative results demonstrate significant improvements in
accuracy, sensitivity, and specificity compared to traditional CNN-based
approaches. This integration of polynomial-based spectral approximation within
deep learning provides a robust framework for enhancing automated medical
diagnostics and holds potential for broader applications in clinical decision
support systems.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 12:02:56 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Roy",
"Abhinav",
""
],
[
"Gyanchandani",
"Bhavesh",
""
],
[
"Oza",
"Aditya",
""
]
] | TITLE: Hybrid CNN with Chebyshev Polynomial Expansion for Medical Image
Analysis
ABSTRACT: Lung cancer remains one of the leading causes of cancer-related mortality
worldwide, with early and accurate diagnosis playing a pivotal role in
improving patient outcomes. Automated detection of pulmonary nodules in
computed tomography (CT) scans is a challenging task due to variability in
nodule size, shape, texture, and location. Traditional Convolutional Neural
Networks (CNNs) have shown considerable promise in medical image analysis;
however, their limited ability to capture fine-grained spatial-spectral
variations restricts their performance in complex diagnostic scenarios. In this
study, we propose a novel hybrid deep learning architecture that incorporates
Chebyshev polynomial expansions into CNN layers to enhance expressive power and
improve the representation of underlying anatomical structures. The proposed
Chebyshev-CNN leverages the orthogonality and recursive properties of Chebyshev
polynomials to extract high-frequency features and approximate complex
nonlinear functions with greater fidelity. The model is trained and evaluated
on benchmark lung cancer imaging datasets, including LUNA16 and LIDC-IDRI,
achieving superior performance in classifying pulmonary nodules as benign or
malignant. Quantitative results demonstrate significant improvements in
accuracy, sensitivity, and specificity compared to traditional CNN-based
approaches. This integration of polynomial-based spectral approximation within
deep learning provides a robust framework for enhancing automated medical
diagnostics and holds potential for broader applications in clinical decision
support systems.
|
2504.06829 | Mahdieh Alizadeh | Ali Goli, Mahdieh Alizadeh, Hadi Sadoghi Yazdi | Adaptive Locally Linear Embedding | 16 pages | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by-sa/4.0/ | Manifold learning techniques, such as Locally linear embedding (LLE), are
designed to preserve the local neighborhood structures of high-dimensional data
during dimensionality reduction. Traditional LLE employs Euclidean distance to
define neighborhoods, which can struggle to capture the intrinsic geometric
relationships within complex data. A novel approach, Adaptive locally linear
embedding(ALLE), is introduced to address this limitation by incorporating a
dynamic, data-driven metric that enhances topological preservation. This method
redefines the concept of proximity by focusing on topological neighborhood
inclusion rather than fixed distances. By adapting the metric based on the
local structure of the data, it achieves superior neighborhood preservation,
particularly for datasets with complex geometries and high-dimensional
structures. Experimental results demonstrate that ALLE significantly improves
the alignment between neighborhoods in the input and feature spaces, resulting
in more accurate and topologically faithful embeddings. This approach advances
manifold learning by tailoring distance metrics to the underlying data,
providing a robust solution for capturing intricate relationships in
high-dimensional datasets.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 12:40:13 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Goli",
"Ali",
""
],
[
"Alizadeh",
"Mahdieh",
""
],
[
"Yazdi",
"Hadi Sadoghi",
""
]
] | TITLE: Adaptive Locally Linear Embedding
ABSTRACT: Manifold learning techniques, such as Locally linear embedding (LLE), are
designed to preserve the local neighborhood structures of high-dimensional data
during dimensionality reduction. Traditional LLE employs Euclidean distance to
define neighborhoods, which can struggle to capture the intrinsic geometric
relationships within complex data. A novel approach, Adaptive locally linear
embedding(ALLE), is introduced to address this limitation by incorporating a
dynamic, data-driven metric that enhances topological preservation. This method
redefines the concept of proximity by focusing on topological neighborhood
inclusion rather than fixed distances. By adapting the metric based on the
local structure of the data, it achieves superior neighborhood preservation,
particularly for datasets with complex geometries and high-dimensional
structures. Experimental results demonstrate that ALLE significantly improves
the alignment between neighborhoods in the input and feature spaces, resulting
in more accurate and topologically faithful embeddings. This approach advances
manifold learning by tailoring distance metrics to the underlying data,
providing a robust solution for capturing intricate relationships in
high-dimensional datasets.
|
2504.06835 | Ziyi Wang | Ziyi Wang, Haoran Wu, Yiming Rong, Deyang Jiang, Yixin Zhang, Yunlong
Zhao, Shuang Xu, Bo XU | LVC: A Lightweight Compression Framework for Enhancing VLMs in Long
Video Understanding | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Long video understanding is a complex task that requires both spatial detail
and temporal awareness. While Vision-Language Models (VLMs) obtain frame-level
understanding capabilities through multi-frame input, they suffer from
information loss due to the sparse sampling strategy. In contrast, Video Large
Language Models (Video-LLMs) capture temporal relationships within visual
features but are limited by the scarcity of high-quality video-text datasets.
To transfer long video understanding capabilities to VLMs with minimal data and
computational cost, we propose Lightweight Video Compression (LVC), a novel
method featuring the Query-Attention Video Compression mechanism, which
effectively tackles the sparse sampling problem in VLMs. By training only the
alignment layer with 10k short video-text pairs, LVC significantly enhances the
temporal reasoning abilities of VLMs. Extensive experiments show that LVC
provides consistent performance improvements across various models, including
the InternVL2 series and Phi-3.5-Vision. Notably, the InternVL2-40B-LVC
achieves scores of 68.2 and 65.9 on the long video understanding benchmarks
MLVU and Video-MME, respectively, with relative improvements of 14.6% and 7.7%.
The enhanced models and code will be publicly available soon.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 12:51:10 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Wang",
"Ziyi",
""
],
[
"Wu",
"Haoran",
""
],
[
"Rong",
"Yiming",
""
],
[
"Jiang",
"Deyang",
""
],
[
"Zhang",
"Yixin",
""
],
[
"Zhao",
"Yunlong",
""
],
[
"Xu",
"Shuang",
""
],
[
"XU",
"Bo",
""
]
] | TITLE: LVC: A Lightweight Compression Framework for Enhancing VLMs in Long
Video Understanding
ABSTRACT: Long video understanding is a complex task that requires both spatial detail
and temporal awareness. While Vision-Language Models (VLMs) obtain frame-level
understanding capabilities through multi-frame input, they suffer from
information loss due to the sparse sampling strategy. In contrast, Video Large
Language Models (Video-LLMs) capture temporal relationships within visual
features but are limited by the scarcity of high-quality video-text datasets.
To transfer long video understanding capabilities to VLMs with minimal data and
computational cost, we propose Lightweight Video Compression (LVC), a novel
method featuring the Query-Attention Video Compression mechanism, which
effectively tackles the sparse sampling problem in VLMs. By training only the
alignment layer with 10k short video-text pairs, LVC significantly enhances the
temporal reasoning abilities of VLMs. Extensive experiments show that LVC
provides consistent performance improvements across various models, including
the InternVL2 series and Phi-3.5-Vision. Notably, the InternVL2-40B-LVC
achieves scores of 68.2 and 65.9 on the long video understanding benchmarks
MLVU and Video-MME, respectively, with relative improvements of 14.6% and 7.7%.
The enhanced models and code will be publicly available soon.
|
2504.06836 | Chun Kit Wong | Jakub Maciej Wi\'sniewski, Anders Nymark Christensen, Mary Le Ngo,
Martin Gr{\o}nneb{\ae}k Tolsgaard, Chun Kit Wong | Determining Fetal Orientations From Blind Sweep Ultrasound Video | 10 pages | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Cognitive demands of fetal ultrasound examinations pose unique challenges
among clinicians. With the goal of providing an assistive tool, we developed an
automated pipeline for predicting fetal orientation from ultrasound videos
acquired following a simple blind sweep protocol. Leveraging on a pre-trained
head detection and segmentation model, this is achieved by first determining
the fetal presentation (cephalic or breech) with a template matching approach,
followed by the fetal lie (facing left or right) by analyzing the spatial
distribution of segmented brain anatomies. Evaluation on a dataset of
third-trimester ultrasound scans demonstrated the promising accuracy of our
pipeline. This work distinguishes itself by introducing automated fetal lie
prediction and by proposing an assistive paradigm that augments sonographer
expertise rather than replacing it. Future research will focus on enhancing
acquisition efficiency, and exploring real-time clinical integration to improve
workflow and support for obstetric clinicians.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 12:51:15 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Wiśniewski",
"Jakub Maciej",
""
],
[
"Christensen",
"Anders Nymark",
""
],
[
"Ngo",
"Mary Le",
""
],
[
"Tolsgaard",
"Martin Grønnebæk",
""
],
[
"Wong",
"Chun Kit",
""
]
] | TITLE: Determining Fetal Orientations From Blind Sweep Ultrasound Video
ABSTRACT: Cognitive demands of fetal ultrasound examinations pose unique challenges
among clinicians. With the goal of providing an assistive tool, we developed an
automated pipeline for predicting fetal orientation from ultrasound videos
acquired following a simple blind sweep protocol. Leveraging on a pre-trained
head detection and segmentation model, this is achieved by first determining
the fetal presentation (cephalic or breech) with a template matching approach,
followed by the fetal lie (facing left or right) by analyzing the spatial
distribution of segmented brain anatomies. Evaluation on a dataset of
third-trimester ultrasound scans demonstrated the promising accuracy of our
pipeline. This work distinguishes itself by introducing automated fetal lie
prediction and by proposing an assistive paradigm that augments sonographer
expertise rather than replacing it. Future research will focus on enhancing
acquisition efficiency, and exploring real-time clinical integration to improve
workflow and support for obstetric clinicians.
|
2504.06841 | Tom Simon | Tom Simon and William Mocaer and Pierrick Tranouez and Clement
Chatelain and Thierry Paquet | Classifying the Unknown: In-Context Learning for Open-Vocabulary Text
and Symbol Recognition | Submitted to ICDAR 2025 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce Rosetta, a multimodal model that leverages Multimodal In-Context
Learning (MICL) to classify sequences of novel script patterns in documents by
leveraging minimal examples, thus eliminating the need for explicit retraining.
To enhance contextual learning, we designed a dataset generation process that
ensures varying degrees of contextual informativeness, improving the model's
adaptability in leveraging context across different scenarios. A key strength
of our method is the use of a Context-Aware Tokenizer (CAT), which enables
open-vocabulary classification. This allows the model to classify text and
symbol patterns across an unlimited range of classes, extending its
classification capabilities beyond the scope of its training alphabet of
patterns. As a result, it unlocks applications such as the recognition of new
alphabets and languages. Experiments on synthetic datasets demonstrate the
potential of Rosetta to successfully classify Out-Of-Distribution visual
patterns and diverse sets of alphabets and scripts, including but not limited
to Chinese, Greek, Russian, French, Spanish, and Japanese.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 12:58:25 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Simon",
"Tom",
""
],
[
"Mocaer",
"William",
""
],
[
"Tranouez",
"Pierrick",
""
],
[
"Chatelain",
"Clement",
""
],
[
"Paquet",
"Thierry",
""
]
] | TITLE: Classifying the Unknown: In-Context Learning for Open-Vocabulary Text
and Symbol Recognition
ABSTRACT: We introduce Rosetta, a multimodal model that leverages Multimodal In-Context
Learning (MICL) to classify sequences of novel script patterns in documents by
leveraging minimal examples, thus eliminating the need for explicit retraining.
To enhance contextual learning, we designed a dataset generation process that
ensures varying degrees of contextual informativeness, improving the model's
adaptability in leveraging context across different scenarios. A key strength
of our method is the use of a Context-Aware Tokenizer (CAT), which enables
open-vocabulary classification. This allows the model to classify text and
symbol patterns across an unlimited range of classes, extending its
classification capabilities beyond the scope of its training alphabet of
patterns. As a result, it unlocks applications such as the recognition of new
alphabets and languages. Experiments on synthetic datasets demonstrate the
potential of Rosetta to successfully classify Out-Of-Distribution visual
patterns and diverse sets of alphabets and scripts, including but not limited
to Chinese, Greek, Russian, French, Spanish, and Japanese.
|
2504.06857 | Roger Huang | Roger G. Huang, Andrew Cudd, Masaki Kawaue, Tatsuya Kikawa, Benjamin
Nachman, Vinicius Mikuni, Callum Wilkinson | Machine Learning-Assisted Unfolding for Neutrino Cross-section
Measurements | 16 pages, 12 figures, 4 tables | null | null | null | hep-ex hep-ph physics.data-an | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The choice of unfolding method for a cross-section measurement is tightly
coupled to the model dependence of the efficiency correction and the overall
impact of cross-section modeling uncertainties in the analysis. A key issue is
the dimensionality used in unfolding, as the kinematics of all outgoing
particles in an event typically affect the reconstruction performance in a
neutrino detector. OmniFold is an unfolding method that iteratively reweights a
simulated dataset, using machine learning to utilize arbitrarily
high-dimensional information, that has previously been applied to proton-proton
and proton-electron datasets. This paper demonstrates OmniFold's application to
a neutrino cross-section measurement for the first time using a public T2K near
detector simulated dataset, comparing its performance with traditional
approaches using a mock data study.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 13:08:35 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Huang",
"Roger G.",
""
],
[
"Cudd",
"Andrew",
""
],
[
"Kawaue",
"Masaki",
""
],
[
"Kikawa",
"Tatsuya",
""
],
[
"Nachman",
"Benjamin",
""
],
[
"Mikuni",
"Vinicius",
""
],
[
"Wilkinson",
"Callum",
""
]
] | TITLE: Machine Learning-Assisted Unfolding for Neutrino Cross-section
Measurements
ABSTRACT: The choice of unfolding method for a cross-section measurement is tightly
coupled to the model dependence of the efficiency correction and the overall
impact of cross-section modeling uncertainties in the analysis. A key issue is
the dimensionality used in unfolding, as the kinematics of all outgoing
particles in an event typically affect the reconstruction performance in a
neutrino detector. OmniFold is an unfolding method that iteratively reweights a
simulated dataset, using machine learning to utilize arbitrarily
high-dimensional information, that has previously been applied to proton-proton
and proton-electron datasets. This paper demonstrates OmniFold's application to
a neutrino cross-section measurement for the first time using a public T2K near
detector simulated dataset, comparing its performance with traditional
approaches using a mock data study.
|
2504.06866 | Seunghyeok Back | Seunghyeok Back, Joosoon Lee, Kangmin Kim, Heeseon Rho, Geonhyup Lee,
Raeyoung Kang, Sangbeom Lee, Sangjun Noh, Youngjin Lee, Taeyeop Lee, Kyoobin
Lee | GraspClutter6D: A Large-scale Real-world Dataset for Robust Perception
and Grasping in Cluttered Scenes | null | null | null | null | cs.RO cs.AI cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Robust grasping in cluttered environments remains an open challenge in
robotics. While benchmark datasets have significantly advanced deep learning
methods, they mainly focus on simplistic scenes with light occlusion and
insufficient diversity, limiting their applicability to practical scenarios. We
present GraspClutter6D, a large-scale real-world grasping dataset featuring:
(1) 1,000 highly cluttered scenes with dense arrangements (14.1 objects/scene,
62.6\% occlusion), (2) comprehensive coverage across 200 objects in 75
environment configurations (bins, shelves, and tables) captured using four
RGB-D cameras from multiple viewpoints, and (3) rich annotations including 736K
6D object poses and 9.3B feasible robotic grasps for 52K RGB-D images. We
benchmark state-of-the-art segmentation, object pose estimation, and grasping
detection methods to provide key insights into challenges in cluttered
environments. Additionally, we validate the dataset's effectiveness as a
training resource, demonstrating that grasping networks trained on
GraspClutter6D significantly outperform those trained on existing datasets in
both simulation and real-world experiments. The dataset, toolkit, and
annotation tools are publicly available on our project website:
https://sites.google.com/view/graspclutter6d.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 13:15:46 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Back",
"Seunghyeok",
""
],
[
"Lee",
"Joosoon",
""
],
[
"Kim",
"Kangmin",
""
],
[
"Rho",
"Heeseon",
""
],
[
"Lee",
"Geonhyup",
""
],
[
"Kang",
"Raeyoung",
""
],
[
"Lee",
"Sangbeom",
""
],
[
"Noh",
"Sangjun",
""
],
[
"Lee",
"Youngjin",
""
],
[
"Lee",
"Taeyeop",
""
],
[
"Lee",
"Kyoobin",
""
]
] | TITLE: GraspClutter6D: A Large-scale Real-world Dataset for Robust Perception
and Grasping in Cluttered Scenes
ABSTRACT: Robust grasping in cluttered environments remains an open challenge in
robotics. While benchmark datasets have significantly advanced deep learning
methods, they mainly focus on simplistic scenes with light occlusion and
insufficient diversity, limiting their applicability to practical scenarios. We
present GraspClutter6D, a large-scale real-world grasping dataset featuring:
(1) 1,000 highly cluttered scenes with dense arrangements (14.1 objects/scene,
62.6\% occlusion), (2) comprehensive coverage across 200 objects in 75
environment configurations (bins, shelves, and tables) captured using four
RGB-D cameras from multiple viewpoints, and (3) rich annotations including 736K
6D object poses and 9.3B feasible robotic grasps for 52K RGB-D images. We
benchmark state-of-the-art segmentation, object pose estimation, and grasping
detection methods to provide key insights into challenges in cluttered
environments. Additionally, we validate the dataset's effectiveness as a
training resource, demonstrating that grasping networks trained on
GraspClutter6D significantly outperform those trained on existing datasets in
both simulation and real-world experiments. The dataset, toolkit, and
annotation tools are publicly available on our project website:
https://sites.google.com/view/graspclutter6d.
|
2504.06880 | Rio Kishimoto | Rio Kishimoto and Tetsuya Kanda and Yuki Manabe and Katsuro Inoue and
Shi Qiu and Yoshiki Higo | A Dataset of Software Bill of Materials for Evaluating SBOM Consumption
Tools | 5 pages, to appear in the Proceedings of the 22nd IEEE/ACM
International Conference on Mining Software Repositories (MSR'25) | null | null | null | cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A Software Bill of Materials (SBOM) is becoming an essential tool for
effective software dependency management. An SBOM is a list of components used
in software, including details such as component names, versions, and licenses.
Using SBOMs, developers can quickly identify software components and assess
whether their software depends on vulnerable libraries. Numerous tools support
software dependency management through SBOMs, which can be broadly categorized
into two types: tools that generate SBOMs and tools that utilize SBOMs. A
substantial collection of accurate SBOMs is required to evaluate tools that
utilize SBOMs. However, there is no publicly available dataset specifically
designed for this purpose, and research on SBOM consumption tools remains
limited. In this paper, we present a dataset of SBOMs to address this gap. The
dataset we constructed comprises 46 SBOMs generated from real-world Java
projects, with plans to expand it to include a broader range of projects across
various programming languages. Accurate and well-structured SBOMs enable
researchers to evaluate the functionality of SBOM consumption tools and
identify potential issues. We collected 3,271 Java projects from GitHub and
generated SBOMs for 798 of them using Maven with an open-source SBOM generation
tool. These SBOMs were refined through both automatic and manual corrections to
ensure accuracy, currently resulting in 46 SBOMs that comply with the SPDX Lite
profile, which defines minimal requirements tailored to practical workflows in
industries. This process also revealed issues with the SBOM generation tools
themselves. The dataset is publicly available on Zenodo (DOI:
10.5281/zenodo.14233415).
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 13:35:02 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Kishimoto",
"Rio",
""
],
[
"Kanda",
"Tetsuya",
""
],
[
"Manabe",
"Yuki",
""
],
[
"Inoue",
"Katsuro",
""
],
[
"Qiu",
"Shi",
""
],
[
"Higo",
"Yoshiki",
""
]
] | TITLE: A Dataset of Software Bill of Materials for Evaluating SBOM Consumption
Tools
ABSTRACT: A Software Bill of Materials (SBOM) is becoming an essential tool for
effective software dependency management. An SBOM is a list of components used
in software, including details such as component names, versions, and licenses.
Using SBOMs, developers can quickly identify software components and assess
whether their software depends on vulnerable libraries. Numerous tools support
software dependency management through SBOMs, which can be broadly categorized
into two types: tools that generate SBOMs and tools that utilize SBOMs. A
substantial collection of accurate SBOMs is required to evaluate tools that
utilize SBOMs. However, there is no publicly available dataset specifically
designed for this purpose, and research on SBOM consumption tools remains
limited. In this paper, we present a dataset of SBOMs to address this gap. The
dataset we constructed comprises 46 SBOMs generated from real-world Java
projects, with plans to expand it to include a broader range of projects across
various programming languages. Accurate and well-structured SBOMs enable
researchers to evaluate the functionality of SBOM consumption tools and
identify potential issues. We collected 3,271 Java projects from GitHub and
generated SBOMs for 798 of them using Maven with an open-source SBOM generation
tool. These SBOMs were refined through both automatic and manual corrections to
ensure accuracy, currently resulting in 46 SBOMs that comply with the SPDX Lite
profile, which defines minimal requirements tailored to practical workflows in
industries. This process also revealed issues with the SBOM generation tools
themselves. The dataset is publicly available on Zenodo (DOI:
10.5281/zenodo.14233415).
|
2504.06881 | Ye Luo | Mingbo Li, Liying Liu, Ye Luo | Compound and Parallel Modes of Tropical Convolutional Neural Networks | 28 pages, 5 figures | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Convolutional neural networks have become increasingly deep and complex,
leading to higher computational costs. While tropical convolutional neural
networks (TCNNs) reduce multiplications, they underperform compared to standard
CNNs. To address this, we propose two new variants - compound TCNN (cTCNN) and
parallel TCNN (pTCNN)-that use combinations of tropical min-plus and max-plus
kernels to replace traditional convolution kernels. This reduces
multiplications and balances efficiency with performance. Experiments on
various datasets show that cTCNN and pTCNN match or exceed the performance of
other CNN methods. Combining these with conventional CNNs in deeper
architectures also improves performance. We are further exploring simplified
TCNN architectures that reduce parameters and multiplications with minimal
accuracy loss, aiming for efficient and effective models.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 13:36:11 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Li",
"Mingbo",
""
],
[
"Liu",
"Liying",
""
],
[
"Luo",
"Ye",
""
]
] | TITLE: Compound and Parallel Modes of Tropical Convolutional Neural Networks
ABSTRACT: Convolutional neural networks have become increasingly deep and complex,
leading to higher computational costs. While tropical convolutional neural
networks (TCNNs) reduce multiplications, they underperform compared to standard
CNNs. To address this, we propose two new variants - compound TCNN (cTCNN) and
parallel TCNN (pTCNN)-that use combinations of tropical min-plus and max-plus
kernels to replace traditional convolution kernels. This reduces
multiplications and balances efficiency with performance. Experiments on
various datasets show that cTCNN and pTCNN match or exceed the performance of
other CNN methods. Combining these with conventional CNNs in deeper
architectures also improves performance. We are further exploring simplified
TCNN architectures that reduce parameters and multiplications with minimal
accuracy loss, aiming for efficient and effective models.
|
2504.06884 | Wuyang Liu | Wuyang Liu, Yi Chai, Yongpeng Yan, Yanzhen Ren | Audio-visual Event Localization on Portrait Mode Short Videos | null | null | null | null | cs.MM cs.AI cs.CV | http://creativecommons.org/licenses/by/4.0/ | Audio-visual event localization (AVEL) plays a critical role in multimodal
scene understanding. While existing datasets for AVEL predominantly comprise
landscape-oriented long videos with clean and simple audio context, short
videos have become the primary format of online video content due to the the
proliferation of smartphones. Short videos are characterized by
portrait-oriented framing and layered audio compositions (e.g., overlapping
sound effects, voiceovers, and music), which brings unique challenges
unaddressed by conventional methods. To this end, we introduce AVE-PM, the
first AVEL dataset specifically designed for portrait mode short videos,
comprising 25,335 clips that span 86 fine-grained categories with frame-level
annotations. Beyond dataset creation, our empirical analysis shows that
state-of-the-art AVEL methods suffer an average 18.66% performance drop during
cross-mode evaluation. Further analysis reveals two key challenges of different
video formats: 1) spatial bias from portrait-oriented framing introduces
distinct domain priors, and 2) noisy audio composition compromise the
reliability of audio modality. To address these issues, we investigate optimal
preprocessing recipes and the impact of background music for AVEL on portrait
mode videos. Experiments show that these methods can still benefit from
tailored preprocessing and specialized model design, thus achieving improved
performance. This work provides both a foundational benchmark and actionable
insights for advancing AVEL research in the era of mobile-centric video
content. Dataset and code will be released.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 13:38:40 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Liu",
"Wuyang",
""
],
[
"Chai",
"Yi",
""
],
[
"Yan",
"Yongpeng",
""
],
[
"Ren",
"Yanzhen",
""
]
] | TITLE: Audio-visual Event Localization on Portrait Mode Short Videos
ABSTRACT: Audio-visual event localization (AVEL) plays a critical role in multimodal
scene understanding. While existing datasets for AVEL predominantly comprise
landscape-oriented long videos with clean and simple audio context, short
videos have become the primary format of online video content due to the the
proliferation of smartphones. Short videos are characterized by
portrait-oriented framing and layered audio compositions (e.g., overlapping
sound effects, voiceovers, and music), which brings unique challenges
unaddressed by conventional methods. To this end, we introduce AVE-PM, the
first AVEL dataset specifically designed for portrait mode short videos,
comprising 25,335 clips that span 86 fine-grained categories with frame-level
annotations. Beyond dataset creation, our empirical analysis shows that
state-of-the-art AVEL methods suffer an average 18.66% performance drop during
cross-mode evaluation. Further analysis reveals two key challenges of different
video formats: 1) spatial bias from portrait-oriented framing introduces
distinct domain priors, and 2) noisy audio composition compromise the
reliability of audio modality. To address these issues, we investigate optimal
preprocessing recipes and the impact of background music for AVEL on portrait
mode videos. Experiments show that these methods can still benefit from
tailored preprocessing and specialized model design, thus achieving improved
performance. This work provides both a foundational benchmark and actionable
insights for advancing AVEL research in the era of mobile-centric video
content. Dataset and code will be released.
|
2504.06908 | Abdullah Hamdi | Emmanuelle Bourigault, Amir Jamaludin, Abdullah Hamdi | UKBOB: One Billion MRI Labeled Masks for Generalizable 3D Medical Image
Segmentation | preprint | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | In medical imaging, the primary challenge is collecting large-scale labeled
data due to privacy concerns, logistics, and high labeling costs. In this work,
we present the UK Biobank Organs and Bones (UKBOB), the largest labeled dataset
of body organs, comprising 51,761 MRI 3D samples (equivalent to 17.9 million 2D
images) and more than 1.37 billion 2D segmentation masks of 72 organs, all
based on the UK Biobank MRI dataset. We utilize automatic labeling, introduce
an automated label cleaning pipeline with organ-specific filters, and manually
annotate a subset of 300 MRIs with 11 abdominal classes to validate the quality
(referred to as UKBOB-manual). This approach allows for scaling up the dataset
collection while maintaining confidence in the labels. We further confirm the
validity of the labels by demonstrating zero-shot generalization of trained
models on the filtered UKBOB to other small labeled datasets from similar
domains (e.g., abdominal MRI). To further mitigate the effect of noisy labels,
we propose a novel method called Entropy Test-time Adaptation (ETTA) to refine
the segmentation output. We use UKBOB to train a foundation model, Swin-BOB,
for 3D medical image segmentation based on the Swin-UNetr architecture,
achieving state-of-the-art results in several benchmarks in 3D medical imaging,
including the BRATS brain MRI tumor challenge (with a 0.4% improvement) and the
BTCV abdominal CT scan benchmark (with a 1.3% improvement). The pre-trained
models and the code are available at https://emmanuelleb985.github.io/ukbob ,
and the filtered labels will be made available with the UK Biobank.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 14:10:51 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Bourigault",
"Emmanuelle",
""
],
[
"Jamaludin",
"Amir",
""
],
[
"Hamdi",
"Abdullah",
""
]
] | TITLE: UKBOB: One Billion MRI Labeled Masks for Generalizable 3D Medical Image
Segmentation
ABSTRACT: In medical imaging, the primary challenge is collecting large-scale labeled
data due to privacy concerns, logistics, and high labeling costs. In this work,
we present the UK Biobank Organs and Bones (UKBOB), the largest labeled dataset
of body organs, comprising 51,761 MRI 3D samples (equivalent to 17.9 million 2D
images) and more than 1.37 billion 2D segmentation masks of 72 organs, all
based on the UK Biobank MRI dataset. We utilize automatic labeling, introduce
an automated label cleaning pipeline with organ-specific filters, and manually
annotate a subset of 300 MRIs with 11 abdominal classes to validate the quality
(referred to as UKBOB-manual). This approach allows for scaling up the dataset
collection while maintaining confidence in the labels. We further confirm the
validity of the labels by demonstrating zero-shot generalization of trained
models on the filtered UKBOB to other small labeled datasets from similar
domains (e.g., abdominal MRI). To further mitigate the effect of noisy labels,
we propose a novel method called Entropy Test-time Adaptation (ETTA) to refine
the segmentation output. We use UKBOB to train a foundation model, Swin-BOB,
for 3D medical image segmentation based on the Swin-UNetr architecture,
achieving state-of-the-art results in several benchmarks in 3D medical imaging,
including the BRATS brain MRI tumor challenge (with a 0.4% improvement) and the
BTCV abdominal CT scan benchmark (with a 1.3% improvement). The pre-trained
models and the code are available at https://emmanuelleb985.github.io/ukbob ,
and the filtered labels will be made available with the UK Biobank.
|
2504.06910 | Sheng Lu | Sheng Lu, Ilia Kuznetsov, Iryna Gurevych | Identifying Aspects in Peer Reviews | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Peer review is central to academic publishing, but the growing volume of
submissions is straining the process. This motivates the development of
computational approaches to support peer review. While each review is tailored
to a specific paper, reviewers often make assessments according to certain
aspects such as Novelty, which reflect the values of the research community.
This alignment creates opportunities for standardizing the reviewing process,
improving quality control, and enabling computational support. While prior work
has demonstrated the potential of aspect analysis for peer review assistance,
the notion of aspect remains poorly formalized. Existing approaches often
derive aspect sets from review forms and guidelines of major NLP venues, yet
data-driven methods for aspect identification are largely underexplored. To
address this gap, our work takes a bottom-up approach: we propose an
operational definition of aspect and develop a data-driven schema for deriving
fine-grained aspects from a corpus of peer reviews. We introduce a dataset of
peer reviews augmented with aspects and show how it can be used for
community-level review analysis. We further show how the choice of aspects can
impact downstream applications, such as LLM-generated review detection. Our
results lay a foundation for a principled and data-driven investigation of
review aspects, and pave the path for new applications of NLP to support peer
review.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 14:14:42 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Lu",
"Sheng",
""
],
[
"Kuznetsov",
"Ilia",
""
],
[
"Gurevych",
"Iryna",
""
]
] | TITLE: Identifying Aspects in Peer Reviews
ABSTRACT: Peer review is central to academic publishing, but the growing volume of
submissions is straining the process. This motivates the development of
computational approaches to support peer review. While each review is tailored
to a specific paper, reviewers often make assessments according to certain
aspects such as Novelty, which reflect the values of the research community.
This alignment creates opportunities for standardizing the reviewing process,
improving quality control, and enabling computational support. While prior work
has demonstrated the potential of aspect analysis for peer review assistance,
the notion of aspect remains poorly formalized. Existing approaches often
derive aspect sets from review forms and guidelines of major NLP venues, yet
data-driven methods for aspect identification are largely underexplored. To
address this gap, our work takes a bottom-up approach: we propose an
operational definition of aspect and develop a data-driven schema for deriving
fine-grained aspects from a corpus of peer reviews. We introduce a dataset of
peer reviews augmented with aspects and show how it can be used for
community-level review analysis. We further show how the choice of aspects can
impact downstream applications, such as LLM-generated review detection. Our
results lay a foundation for a principled and data-driven investigation of
review aspects, and pave the path for new applications of NLP to support peer
review.
|
2504.06915 | Francisco Mena | Miro Miranda, Francisco Mena, Andreas Dengel | An Analysis of Temporal Dropout in Earth Observation Time Series for
Regression Tasks | Accepted at Symposium on Intelligent Data Analysis (IDA 2025) | null | null | null | cs.LG cs.AI cs.CV | http://creativecommons.org/licenses/by/4.0/ | Missing instances in time series data impose a significant challenge to deep
learning models, particularly in regression tasks. In the Earth Observation
field, satellite failure or cloud occlusion frequently results in missing
time-steps, introducing uncertainties in the predicted output and causing a
decline in predictive performance. While many studies address missing
time-steps through data augmentation to improve model robustness, the
uncertainty arising at the input level is commonly overlooked. To address this
gap, we introduce Monte Carlo Temporal Dropout (MC-TD), a method that
explicitly accounts for input-level uncertainty by randomly dropping time-steps
during inference using a predefined dropout ratio, thereby simulating the
effect of missing data. To bypass the need for costly searches for the optimal
dropout ratio, we extend this approach with Monte Carlo Concrete Temporal
Dropout (MC-ConcTD), a method that learns the optimal dropout distribution
directly. Both MC-TD and MC-ConcTD are applied during inference, leveraging
Monte Carlo sampling for uncertainty quantification. Experiments on three EO
time-series datasets demonstrate that MC-ConcTD improves predictive performance
and uncertainty calibration compared to existing approaches. Additionally, we
highlight the advantages of adaptive dropout tuning over manual selection,
making uncertainty quantification more robust and accessible for EO
applications.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 14:23:04 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Miranda",
"Miro",
""
],
[
"Mena",
"Francisco",
""
],
[
"Dengel",
"Andreas",
""
]
] | TITLE: An Analysis of Temporal Dropout in Earth Observation Time Series for
Regression Tasks
ABSTRACT: Missing instances in time series data impose a significant challenge to deep
learning models, particularly in regression tasks. In the Earth Observation
field, satellite failure or cloud occlusion frequently results in missing
time-steps, introducing uncertainties in the predicted output and causing a
decline in predictive performance. While many studies address missing
time-steps through data augmentation to improve model robustness, the
uncertainty arising at the input level is commonly overlooked. To address this
gap, we introduce Monte Carlo Temporal Dropout (MC-TD), a method that
explicitly accounts for input-level uncertainty by randomly dropping time-steps
during inference using a predefined dropout ratio, thereby simulating the
effect of missing data. To bypass the need for costly searches for the optimal
dropout ratio, we extend this approach with Monte Carlo Concrete Temporal
Dropout (MC-ConcTD), a method that learns the optimal dropout distribution
directly. Both MC-TD and MC-ConcTD are applied during inference, leveraging
Monte Carlo sampling for uncertainty quantification. Experiments on three EO
time-series datasets demonstrate that MC-ConcTD improves predictive performance
and uncertainty calibration compared to existing approaches. Additionally, we
highlight the advantages of adaptive dropout tuning over manual selection,
making uncertainty quantification more robust and accessible for EO
applications.
|
2504.06917 | Ming Liu | Ming Liu and Massimo Poesio | Data Augmentation for Fake Reviews Detection in Multiple Languages and
Multiple Domains | 32 pages, 15 figures | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | With the growth of the Internet, buying habits have changed, and customers
have become more dependent on the online opinions of other customers to guide
their purchases. Identifying fake reviews thus became an important area for
Natural Language Processing (NLP) research. However, developing
high-performance NLP models depends on the availability of large amounts of
training data, which are often not available for low-resource languages or
domains. In this research, we used large language models to generate datasets
to train fake review detectors. Our approach was used to generate fake reviews
in different domains (book reviews, restaurant reviews, and hotel reviews) and
different languages (English and Chinese). Our results demonstrate that our
data augmentation techniques result in improved performance at fake review
detection for all domains and languages. The accuracy of our fake review
detection model can be improved by 0.3 percentage points on DeRev TEST, 10.9
percentage points on Amazon TEST, 8.3 percentage points on Yelp TEST and 7.2
percentage points on DianPing TEST using the augmented datasets.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 14:23:54 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Liu",
"Ming",
""
],
[
"Poesio",
"Massimo",
""
]
] | TITLE: Data Augmentation for Fake Reviews Detection in Multiple Languages and
Multiple Domains
ABSTRACT: With the growth of the Internet, buying habits have changed, and customers
have become more dependent on the online opinions of other customers to guide
their purchases. Identifying fake reviews thus became an important area for
Natural Language Processing (NLP) research. However, developing
high-performance NLP models depends on the availability of large amounts of
training data, which are often not available for low-resource languages or
domains. In this research, we used large language models to generate datasets
to train fake review detectors. Our approach was used to generate fake reviews
in different domains (book reviews, restaurant reviews, and hotel reviews) and
different languages (English and Chinese). Our results demonstrate that our
data augmentation techniques result in improved performance at fake review
detection for all domains and languages. The accuracy of our fake review
detection model can be improved by 0.3 percentage points on DeRev TEST, 10.9
percentage points on Amazon TEST, 8.3 percentage points on Yelp TEST and 7.2
percentage points on DianPing TEST using the augmented datasets.
|
2504.06920 | Thibaud Ehret | Masquil El\'ias, Mar\'i Roger, Ehret Thibaud, Meinhardt-Llopis Enric,
Mus\'e Pablo, Facciolo Gabriele | S-EO: A Large-Scale Dataset for Geometry-Aware Shadow Detection in
Remote Sensing Applications | Accepted at Earthvision 2025 (CVPR Workshop) | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We introduce the S-EO dataset: a large-scale, high-resolution dataset,
designed to advance geometry-aware shadow detection. Collected from diverse
public-domain sources, including challenge datasets and government providers
such as USGS, our dataset comprises 702 georeferenced tiles across the USA,
each covering 500x500 m. Each tile includes multi-date, multi-angle WorldView-3
pansharpened RGB images, panchromatic images, and a ground-truth DSM of the
area obtained from LiDAR scans. For each image, we provide a shadow mask
derived from geometry and sun position, a vegetation mask based on the NDVI
index, and a bundle-adjusted RPC model. With approximately 20,000 images, the
S-EO dataset establishes a new public resource for shadow detection in remote
sensing imagery and its applications to 3D reconstruction. To demonstrate the
dataset's impact, we train and evaluate a shadow detector, showcasing its
ability to generalize, even to aerial images. Finally, we extend EO-NeRF - a
state-of-the-art NeRF approach for satellite imagery - to leverage our shadow
predictions for improved 3D reconstructions.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 14:25:35 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Elías",
"Masquil",
""
],
[
"Roger",
"Marí",
""
],
[
"Thibaud",
"Ehret",
""
],
[
"Enric",
"Meinhardt-Llopis",
""
],
[
"Pablo",
"Musé",
""
],
[
"Gabriele",
"Facciolo",
""
]
] | TITLE: S-EO: A Large-Scale Dataset for Geometry-Aware Shadow Detection in
Remote Sensing Applications
ABSTRACT: We introduce the S-EO dataset: a large-scale, high-resolution dataset,
designed to advance geometry-aware shadow detection. Collected from diverse
public-domain sources, including challenge datasets and government providers
such as USGS, our dataset comprises 702 georeferenced tiles across the USA,
each covering 500x500 m. Each tile includes multi-date, multi-angle WorldView-3
pansharpened RGB images, panchromatic images, and a ground-truth DSM of the
area obtained from LiDAR scans. For each image, we provide a shadow mask
derived from geometry and sun position, a vegetation mask based on the NDVI
index, and a bundle-adjusted RPC model. With approximately 20,000 images, the
S-EO dataset establishes a new public resource for shadow detection in remote
sensing imagery and its applications to 3D reconstruction. To demonstrate the
dataset's impact, we train and evaluate a shadow detector, showcasing its
ability to generalize, even to aerial images. Finally, we extend EO-NeRF - a
state-of-the-art NeRF approach for satellite imagery - to leverage our shadow
predictions for improved 3D reconstructions.
|
2504.06921 | Tejas Sudharshan Mathai | Anisa V. Prasad, Tejas Sudharshan Mathai, Pritam Mukherjee, Jianfei
Liu, and Ronald M. Summers | Leveraging Anatomical Priors for Automated Pancreas Segmentation on
Abdominal CT | Published at SPIE Medical Imaging 2025 | null | null | null | eess.IV cs.AI cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | An accurate segmentation of the pancreas on CT is crucial to identify
pancreatic pathologies and extract imaging-based biomarkers. However, prior
research on pancreas segmentation has primarily focused on modifying the
segmentation model architecture or utilizing pre- and post-processing
techniques. In this article, we investigate the utility of anatomical priors to
enhance the segmentation performance of the pancreas. Two 3D full-resolution
nnU-Net models were trained, one with 8 refined labels from the public PANORAMA
dataset, and another that combined them with labels derived from the public
TotalSegmentator (TS) tool. The addition of anatomical priors resulted in a 6\%
increase in Dice score ($p < .001$) and a 36.5 mm decrease in Hausdorff
distance for pancreas segmentation ($p < .001$). Moreover, the pancreas was
always detected when anatomy priors were used, whereas there were 8 instances
of failed detections without their use. The use of anatomy priors shows promise
for pancreas segmentation and subsequent derivation of imaging biomarkers.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 14:29:08 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Prasad",
"Anisa V.",
""
],
[
"Mathai",
"Tejas Sudharshan",
""
],
[
"Mukherjee",
"Pritam",
""
],
[
"Liu",
"Jianfei",
""
],
[
"Summers",
"Ronald M.",
""
]
] | TITLE: Leveraging Anatomical Priors for Automated Pancreas Segmentation on
Abdominal CT
ABSTRACT: An accurate segmentation of the pancreas on CT is crucial to identify
pancreatic pathologies and extract imaging-based biomarkers. However, prior
research on pancreas segmentation has primarily focused on modifying the
segmentation model architecture or utilizing pre- and post-processing
techniques. In this article, we investigate the utility of anatomical priors to
enhance the segmentation performance of the pancreas. Two 3D full-resolution
nnU-Net models were trained, one with 8 refined labels from the public PANORAMA
dataset, and another that combined them with labels derived from the public
TotalSegmentator (TS) tool. The addition of anatomical priors resulted in a 6\%
increase in Dice score ($p < .001$) and a 36.5 mm decrease in Hausdorff
distance for pancreas segmentation ($p < .001$). Moreover, the pancreas was
always detected when anatomy priors were used, whereas there were 8 instances
of failed detections without their use. The use of anatomy priors shows promise
for pancreas segmentation and subsequent derivation of imaging biomarkers.
|
2504.06923 | Emiliano De Cristofaro | Georgi Ganev and Meenatchi Sundaram Muthu Selva Annamalai and Sofiane
Mahiou and Emiliano De Cristofaro | The Importance of Being Discrete: Measuring the Impact of Discretization
in End-to-End Differentially Private Synthetic Data | null | null | null | null | cs.CR cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Differentially Private (DP) generative marginal models are often used in the
wild to release synthetic tabular datasets in lieu of sensitive data while
providing formal privacy guarantees. These models approximate low-dimensional
marginals or query workloads; crucially, they require the training data to be
pre-discretized, i.e., continuous values need to first be partitioned into
bins. However, as the range of values (or their domain) is often inferred
directly from the training data, with the number of bins and bin edges
typically defined arbitrarily, this approach can ultimately break end-to-end DP
guarantees and may not always yield optimal utility.
In this paper, we present an extensive measurement study of four
discretization strategies in the context of DP marginal generative models. More
precisely, we design DP versions of three discretizers (uniform, quantile, and
k-means) and reimplement the PrivTree algorithm. We find that optimizing both
the choice of discretizer and bin count can improve utility, on average, by
almost 30% across six DP marginal models, compared to the default strategy and
number of bins, with PrivTree being the best-performing discretizer in the
majority of cases. We demonstrate that, while DP generative models with
non-private discretization remain vulnerable to membership inference attacks,
applying DP during discretization effectively mitigates this risk. Finally, we
propose an optimized approach for automatically selecting the optimal number of
bins, achieving high utility while reducing both privacy budget consumption and
computational overhead.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 14:30:30 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Ganev",
"Georgi",
""
],
[
"Annamalai",
"Meenatchi Sundaram Muthu Selva",
""
],
[
"Mahiou",
"Sofiane",
""
],
[
"De Cristofaro",
"Emiliano",
""
]
] | TITLE: The Importance of Being Discrete: Measuring the Impact of Discretization
in End-to-End Differentially Private Synthetic Data
ABSTRACT: Differentially Private (DP) generative marginal models are often used in the
wild to release synthetic tabular datasets in lieu of sensitive data while
providing formal privacy guarantees. These models approximate low-dimensional
marginals or query workloads; crucially, they require the training data to be
pre-discretized, i.e., continuous values need to first be partitioned into
bins. However, as the range of values (or their domain) is often inferred
directly from the training data, with the number of bins and bin edges
typically defined arbitrarily, this approach can ultimately break end-to-end DP
guarantees and may not always yield optimal utility.
In this paper, we present an extensive measurement study of four
discretization strategies in the context of DP marginal generative models. More
precisely, we design DP versions of three discretizers (uniform, quantile, and
k-means) and reimplement the PrivTree algorithm. We find that optimizing both
the choice of discretizer and bin count can improve utility, on average, by
almost 30% across six DP marginal models, compared to the default strategy and
number of bins, with PrivTree being the best-performing discretizer in the
majority of cases. We demonstrate that, while DP generative models with
non-private discretization remain vulnerable to membership inference attacks,
applying DP during discretization effectively mitigates this risk. Finally, we
propose an optimized approach for automatically selecting the optimal number of
bins, achieving high utility while reducing both privacy budget consumption and
computational overhead.
|
2504.06927 | Ur\v{s}ka Matja\v{s}ec | Ur\v{s}ka Matja\v{s}ec, Nikola Simidjievski, Mateja Jamnik | RO-FIGS: Efficient and Expressive Tree-Based Ensembles for Tabular Data | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Tree-based models are often robust to uninformative features and can
accurately capture non-smooth, complex decision boundaries. Consequently, they
often outperform neural network-based models on tabular datasets at a
significantly lower computational cost. Nevertheless, the capability of
traditional tree-based ensembles to express complex relationships efficiently
is limited by using a single feature to make splits. To improve the efficiency
and expressiveness of tree-based methods, we propose Random Oblique Fast
Interpretable Greedy-Tree Sums (RO-FIGS). RO-FIGS builds on Fast Interpretable
Greedy-Tree Sums, and extends it by learning trees with oblique or multivariate
splits, where each split consists of a linear combination learnt from random
subsets of features. This helps uncover interactions between features and
improves performance. The proposed method is suitable for tabular datasets with
both numerical and categorical features. We evaluate RO-FIGS on 22 real-world
tabular datasets, demonstrating superior performance and much smaller models
over other tree- and neural network-based methods. Additionally, we analyse
their splits to reveal valuable insights into feature interactions, enriching
the information learnt from SHAP summary plots, and thereby demonstrating the
enhanced interpretability of RO-FIGS models. The proposed method is well-suited
for applications, where balance between accuracy and interpretability is
essential.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 14:35:24 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Matjašec",
"Urška",
""
],
[
"Simidjievski",
"Nikola",
""
],
[
"Jamnik",
"Mateja",
""
]
] | TITLE: RO-FIGS: Efficient and Expressive Tree-Based Ensembles for Tabular Data
ABSTRACT: Tree-based models are often robust to uninformative features and can
accurately capture non-smooth, complex decision boundaries. Consequently, they
often outperform neural network-based models on tabular datasets at a
significantly lower computational cost. Nevertheless, the capability of
traditional tree-based ensembles to express complex relationships efficiently
is limited by using a single feature to make splits. To improve the efficiency
and expressiveness of tree-based methods, we propose Random Oblique Fast
Interpretable Greedy-Tree Sums (RO-FIGS). RO-FIGS builds on Fast Interpretable
Greedy-Tree Sums, and extends it by learning trees with oblique or multivariate
splits, where each split consists of a linear combination learnt from random
subsets of features. This helps uncover interactions between features and
improves performance. The proposed method is suitable for tabular datasets with
both numerical and categorical features. We evaluate RO-FIGS on 22 real-world
tabular datasets, demonstrating superior performance and much smaller models
over other tree- and neural network-based methods. Additionally, we analyse
their splits to reveal valuable insights into feature interactions, enriching
the information learnt from SHAP summary plots, and thereby demonstrating the
enhanced interpretability of RO-FIGS models. The proposed method is well-suited
for applications, where balance between accuracy and interpretability is
essential.
|
2504.06935 | Chenyu Hui | Chenyu Hui and Anran Zhang and Xintong Li | ASRL:A robust loss function with potential for development | five pages and three figures | null | null | null | cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | In this article, we proposed a partition:wise robust loss function based on
the previous robust loss function. The characteristics of this loss function
are that it achieves high robustness and a wide range of applicability through
partition-wise design and adaptive parameter adjustment. Finally, the
advantages and development potential of this loss function were verified by
applying this loss function to the regression question and using five different
datasets (with different dimensions, different sample numbers, and different
fields) to compare with the other loss functions. The results of multiple
experiments have proven the advantages of our loss function .
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 14:40:46 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Hui",
"Chenyu",
""
],
[
"Zhang",
"Anran",
""
],
[
"Li",
"Xintong",
""
]
] | TITLE: ASRL:A robust loss function with potential for development
ABSTRACT: In this article, we proposed a partition:wise robust loss function based on
the previous robust loss function. The characteristics of this loss function
are that it achieves high robustness and a wide range of applicability through
partition-wise design and adaptive parameter adjustment. Finally, the
advantages and development potential of this loss function were verified by
applying this loss function to the regression question and using five different
datasets (with different dimensions, different sample numbers, and different
fields) to compare with the other loss functions. The results of multiple
experiments have proven the advantages of our loss function .
|
2504.06950 | Sachin Kumar Danisetty | Sachin Kumar Danisetty, Alexandros Graikos, Srikar Yellapragada,
Dimitris Samaras | PathSegDiff: Pathology Segmentation using Diffusion model
representations | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Image segmentation is crucial in many computational pathology pipelines,
including accurate disease diagnosis, subtyping, outcome, and survivability
prediction. The common approach for training a segmentation model relies on a
pre-trained feature extractor and a dataset of paired image and mask
annotations. These are used to train a lightweight prediction model that
translates features into per-pixel classes. The choice of the feature extractor
is central to the performance of the final segmentation model, and recent
literature has focused on finding tasks to pre-train the feature extractor. In
this paper, we propose PathSegDiff, a novel approach for histopathology image
segmentation that leverages Latent Diffusion Models (LDMs) as pre-trained
featured extractors. Our method utilizes a pathology-specific LDM, guided by a
self-supervised encoder, to extract rich semantic information from H\&E stained
histopathology images. We employ a simple, fully convolutional network to
process the features extracted from the LDM and generate segmentation masks.
Our experiments demonstrate significant improvements over traditional methods
on the BCSS and GlaS datasets, highlighting the effectiveness of
domain-specific diffusion pre-training in capturing intricate tissue structures
and enhancing segmentation accuracy in histopathology images.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 14:58:21 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Danisetty",
"Sachin Kumar",
""
],
[
"Graikos",
"Alexandros",
""
],
[
"Yellapragada",
"Srikar",
""
],
[
"Samaras",
"Dimitris",
""
]
] | TITLE: PathSegDiff: Pathology Segmentation using Diffusion model
representations
ABSTRACT: Image segmentation is crucial in many computational pathology pipelines,
including accurate disease diagnosis, subtyping, outcome, and survivability
prediction. The common approach for training a segmentation model relies on a
pre-trained feature extractor and a dataset of paired image and mask
annotations. These are used to train a lightweight prediction model that
translates features into per-pixel classes. The choice of the feature extractor
is central to the performance of the final segmentation model, and recent
literature has focused on finding tasks to pre-train the feature extractor. In
this paper, we propose PathSegDiff, a novel approach for histopathology image
segmentation that leverages Latent Diffusion Models (LDMs) as pre-trained
featured extractors. Our method utilizes a pathology-specific LDM, guided by a
self-supervised encoder, to extract rich semantic information from H\&E stained
histopathology images. We employ a simple, fully convolutional network to
process the features extracted from the LDM and generate segmentation masks.
Our experiments demonstrate significant improvements over traditional methods
on the BCSS and GlaS datasets, highlighting the effectiveness of
domain-specific diffusion pre-training in capturing intricate tissue structures
and enhancing segmentation accuracy in histopathology images.
|
2504.06957 | Marco Acerbis | Marco Acerbis, Nata\v{s}a Sladoje, Joakim Lindblad | A Comparison of Deep Learning Methods for Cell Detection in Digital
Cytology | 14 pages, 6 figures, SCIA2025 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Accurate and efficient cell detection is crucial in many biomedical image
analysis tasks. We evaluate the performance of several Deep Learning (DL)
methods for cell detection in Papanicolaou-stained cytological Whole Slide
Images (WSIs), focusing on accuracy of predictions and computational
efficiency. We examine recentoff-the-shelf algorithms as well as
custom-designed detectors, applying them to two datasets: the CNSeg Dataset and
the Oral Cancer (OC) Dataset. Our comparison includes well-established
segmentation methods such as StarDist, Cellpose, and the Segment Anything Model
2 (SAM2), alongside centroid-based Fully Convolutional Regression Network
(FCRN) approaches. We introduce a suitable evaluation metric to assess the
accuracy of predictions based on the distance from ground truth positions. We
also explore the impact of dataset size and data augmentation techniques on
model performance. Results show that centroid-based methods, particularly the
Improved Fully Convolutional Regression Network (IFCRN) method, outperform
segmentation-based methods in terms of both detection accuracy and
computational efficiency. This study highlights the potential of centroid-based
detectors as a preferred option for cell detection in resource-limited
environments, offering faster processing times and lower GPU memory usage
without compromising accuracy.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 15:08:12 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Acerbis",
"Marco",
""
],
[
"Sladoje",
"Nataša",
""
],
[
"Lindblad",
"Joakim",
""
]
] | TITLE: A Comparison of Deep Learning Methods for Cell Detection in Digital
Cytology
ABSTRACT: Accurate and efficient cell detection is crucial in many biomedical image
analysis tasks. We evaluate the performance of several Deep Learning (DL)
methods for cell detection in Papanicolaou-stained cytological Whole Slide
Images (WSIs), focusing on accuracy of predictions and computational
efficiency. We examine recentoff-the-shelf algorithms as well as
custom-designed detectors, applying them to two datasets: the CNSeg Dataset and
the Oral Cancer (OC) Dataset. Our comparison includes well-established
segmentation methods such as StarDist, Cellpose, and the Segment Anything Model
2 (SAM2), alongside centroid-based Fully Convolutional Regression Network
(FCRN) approaches. We introduce a suitable evaluation metric to assess the
accuracy of predictions based on the distance from ground truth positions. We
also explore the impact of dataset size and data augmentation techniques on
model performance. Results show that centroid-based methods, particularly the
Improved Fully Convolutional Regression Network (IFCRN) method, outperform
segmentation-based methods in terms of both detection accuracy and
computational efficiency. This study highlights the potential of centroid-based
detectors as a preferred option for cell detection in resource-limited
environments, offering faster processing times and lower GPU memory usage
without compromising accuracy.
|
2504.06961 | Yu Qi | Yu Qi, Yuanchen Ju, Tianming Wei, Chi Chu, Lawson L.S. Wong, Huazhe Xu | Two by Two: Learning Multi-Task Pairwise Objects Assembly for
Generalizable Robot Manipulation | Accepted to CVPR 2025 (Conference on Computer Vision and Pattern
Recognition) | null | null | null | cs.RO cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | 3D assembly tasks, such as furniture assembly and component fitting, play a
crucial role in daily life and represent essential capabilities for future home
robots. Existing benchmarks and datasets predominantly focus on assembling
geometric fragments or factory parts, which fall short in addressing the
complexities of everyday object interactions and assemblies. To bridge this
gap, we present 2BY2, a large-scale annotated dataset for daily pairwise
objects assembly, covering 18 fine-grained tasks that reflect real-life
scenarios, such as plugging into sockets, arranging flowers in vases, and
inserting bread into toasters. 2BY2 dataset includes 1,034 instances and 517
pairwise objects with pose and symmetry annotations, requiring approaches that
align geometric shapes while accounting for functional and spatial
relationships between objects. Leveraging the 2BY2 dataset, we propose a
two-step SE(3) pose estimation method with equivariant features for assembly
constraints. Compared to previous shape assembly methods, our approach achieves
state-of-the-art performance across all 18 tasks in the 2BY2 dataset.
Additionally, robot experiments further validate the reliability and
generalization ability of our method for complex 3D assembly tasks.
| [
{
"version": "v1",
"created": "Wed, 9 Apr 2025 15:12:38 GMT"
}
] | 2025-04-10T00:00:00 | [
[
"Qi",
"Yu",
""
],
[
"Ju",
"Yuanchen",
""
],
[
"Wei",
"Tianming",
""
],
[
"Chu",
"Chi",
""
],
[
"Wong",
"Lawson L. S.",
""
],
[
"Xu",
"Huazhe",
""
]
] | TITLE: Two by Two: Learning Multi-Task Pairwise Objects Assembly for
Generalizable Robot Manipulation
ABSTRACT: 3D assembly tasks, such as furniture assembly and component fitting, play a
crucial role in daily life and represent essential capabilities for future home
robots. Existing benchmarks and datasets predominantly focus on assembling
geometric fragments or factory parts, which fall short in addressing the
complexities of everyday object interactions and assemblies. To bridge this
gap, we present 2BY2, a large-scale annotated dataset for daily pairwise
objects assembly, covering 18 fine-grained tasks that reflect real-life
scenarios, such as plugging into sockets, arranging flowers in vases, and
inserting bread into toasters. 2BY2 dataset includes 1,034 instances and 517
pairwise objects with pose and symmetry annotations, requiring approaches that
align geometric shapes while accounting for functional and spatial
relationships between objects. Leveraging the 2BY2 dataset, we propose a
two-step SE(3) pose estimation method with equivariant features for assembly
constraints. Compared to previous shape assembly methods, our approach achieves
state-of-the-art performance across all 18 tasks in the 2BY2 dataset.
Additionally, robot experiments further validate the reliability and
generalization ability of our method for complex 3D assembly tasks.
|
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.