app.py

import gradio as gr
import spaces
import torch
from transformers import AutoConfig, AutoTokenizer, AutoModelForSequenceClassification
import torch.nn.functional as F
import torch.nn as nn
import re
import requests
from urllib.parse import urlparse
import xml.etree.ElementTree as ET

##################################################
# Global setup
##################################################
model_path = "ssocean/NAIP"
device = "cuda" if torch.cuda.is_available() else "cpu"

model = None
tokenizer = None

##################################################
# Fetch paper info from arXiv
##################################################
def fetch_arxiv_paper(arxiv_input):
    """
    Fetch paper title & abstract from an arXiv URL or ID.
    """
    try:
        if "arxiv.org" in arxiv_input:
            parsed = urlparse(arxiv_input)
            path = parsed.path
            arxiv_id = path.split("/")[-1].replace(".pdf", "")
        else:
            arxiv_id = arxiv_input.strip()

        api_url = f"http://export.arxiv.org/api/query?id_list={arxiv_id}"
        resp = requests.get(api_url)
        if resp.status_code != 200:
            return {
                "title": "",
                "abstract": "",
                "success": False,
                "message": "Error fetching paper from arXiv API",
            }

        root = ET.fromstring(resp.text)
        ns = {"arxiv": "http://www.w3.org/2005/Atom"}
        entry = root.find(".//arxiv:entry", ns)
        if entry is None:
            return {"title": "", "abstract": "", "success": False, "message": "Paper not found"}

        title = entry.find("arxiv:title", ns).text.strip()
        abstract = entry.find("arxiv:summary", ns).text.strip()

        return {
            "title": title,
            "abstract": abstract,
            "success": True,
            "message": "Paper fetched successfully!",
        }
    except Exception as e:
        return {
            "title": "",
            "abstract": "",
            "success": False,
            "message": f"Error fetching paper: {e}",
        }

##################################################
# Prediction function
##################################################
@spaces.GPU(duration=60, enable_queue=True)
def predict(title, abstract):
    """
    Predict a normalized academic impact score (0–1) from title & abstract.
    """
    global model, tokenizer
    if model is None:
        # 1) Load config
        config = AutoConfig.from_pretrained(model_path)
        
        # 2) Remove quantization_config if it exists (avoid NoneType error in PEFT)
        if hasattr(config, "quantization_config"):
            del config.quantization_config

        # 3) Optionally set number of labels
        config.num_labels = 1

        # 4) Load the model
        model_loaded = AutoModelForSequenceClassification.from_pretrained(
            model_path,
            config=config,
            torch_dtype=torch.float32,  # float32 for stable cublasLt
            device_map=None,
            low_cpu_mem_usage=False
        )
        model_loaded.to(device)
        model_loaded.eval()

        # 5) Load tokenizer
        tokenizer_loaded = AutoTokenizer.from_pretrained(model_path)

        # Assign to globals
        model, tokenizer = model_loaded, tokenizer_loaded

    text = (
        f"Given a certain paper,\n"
        f"Title: {title.strip()}\n"
        f"Abstract: {abstract.strip()}\n"
        f"Predict its normalized academic impact (0~1):"
    )

    try:
        inputs = tokenizer(text, return_tensors="pt", truncation=True, max_length=1024)
        inputs = {k: v.to(device) for k, v in inputs.items()}
        with torch.no_grad():
            outputs = model(**inputs)
        logits = outputs.logits
        prob = torch.sigmoid(logits).item()
        score = min(1.0, prob + 0.05)
        return round(score, 4)
    except Exception as e:
        print("Prediction error:", e)
        return 0.0

##################################################
# Grading
##################################################
def get_grade_and_emoji(score):
    """Map a 0–1 score to an A/B/C style grade with an emoji indicator."""
    if score >= 0.900:
        return "AAA 🌟"
    if score >= 0.800:
        return "AA ⭐"
    if score >= 0.650:
        return "A ✨"
    if score >= 0.600:
        return "BBB 🔵"
    if score >= 0.550:
        return "BB 📘"
    if score >= 0.500:
        return "B 📖"
    if score >= 0.400:
        return "CCC 📝"
    if score >= 0.300:
        return "CC ✏️"
    return "C 📑"

##################################################
# Validation
##################################################
def validate_input(title, abstract):
    """
    Ensure the title has at least 3 words, the abstract at least 50,
    and check for ASCII-only characters.
    """
    non_ascii = re.compile(r"[^\x00-\x7F]")
    if len(title.split()) < 3:
        return False, "Title must be at least 3 words."
    if len(abstract.split()) < 50:
        return False, "Abstract must be at least 50 words."
    if non_ascii.search(title):
        return False, "Title contains non-ASCII characters."
    if non_ascii.search(abstract):
        return False, "Abstract contains non-ASCII characters."
    return True, "Inputs look good."

def update_button_status(title, abstract):
    """Enable or disable the predict button based on validation."""
    valid, msg = validate_input(title, abstract)
    if not valid:
        return gr.update(value="Error: " + msg), gr.update(interactive=False)
    return gr.update(value=msg), gr.update(interactive=True)

##################################################
# Process arXiv input
##################################################
def process_arxiv_input(arxiv_input):
    """
    Called when user clicks 'Fetch Paper Details' to fill in title/abstract from arXiv.
    """
    if not arxiv_input.strip():
        return "", "", "Please enter an arXiv URL or ID"
    res = fetch_arxiv_paper(arxiv_input)
    if res["success"]:
        return res["title"], res["abstract"], res["message"]
    return "", "", res["message"]

##################################################
# Custom CSS
##################################################
css = """
.gradio-container { font-family: Arial, sans-serif; }
.main-title {
    text-align: center; color: #2563eb; font-size: 2.5rem!important;
    margin-bottom:1rem!important;
    background: linear-gradient(45deg,#2563eb,#1d4ed8);
    -webkit-background-clip: text; -webkit-text-fill-color: transparent;
}
.input-section {
    background:#fff; padding:1.5rem; border-radius:0.5rem;
    box-shadow:0 4px 6px rgba(0,0,0,0.1);
}
.result-section {
    background:#f7f9fc; padding:1.5rem; border-radius:0.5rem;
    margin-top:2rem;
}
.grade-display {
    font-size:2.5rem; text-align:center; margin-top:1rem;
}
.arxiv-input {
    margin-bottom:1.5rem; padding:1rem; background:#f3f4f6;
    border-radius:0.5rem;
}
.arxiv-link {
    color:#2563eb; text-decoration: underline;
}
"""

##################################################
# Header HTML (social links)
##################################################
header_html = """
<div align="center" style="line-height: 1;">
  <a href="https://discord.gg/openfreeai" style="margin: 2px;">
    <img alt="OpenFree AI Discord Server" src="https://img.shields.io/badge/Discord-000000?style=for-the-badge&logo=discord&logoColor=000&logoColor=white" style="display: inline-block; vertical-align: middle;"/>
  </a>
  <a href="https://x.com/openfree_ai" style="margin: 2px;">
    <img alt="X.ai" src="https://img.shields.io/badge/openfree_ai-000000?style=for-the-badge&logo=X&logoColor=000&color=000&labelColor=white" style="display: inline-block; vertical-align: middle;"/>
  </a>
  <a href="https://huggingface.co./collections/VIDraft/best-open-ai-services-68057e6e312880ea92abaf4c" style="margin: 2px;">
    <img alt="Collections" src="https://img.shields.io/badge/Collections-f000000?style=for-the-badge&logo=huggingface&logoColor=000&labelColor" style="display: inline-block; vertical-align: middle;"/>
  </a>  
  <a href="https://huggingface.co./VIDraft" style="margin: 2px;">
    <img alt="HF Page" src="https://img.shields.io/badge/VIDraft-fcd022?style=for-the-badge&logo=huggingface&logoColor=000&labelColor" style="display: inline-block; vertical-align: middle;"/>
  </a>  
</div>

<div align="center" style="font-weight: bold; margin-top: 10px; margin-bottom: 15px;">
  <b>Papers Leaderboard: <a href="https://huggingface.co./spaces/Heartsync/Papers-Leaderboard">https://huggingface.co./spaces/Heartsync/Papers-Leaderboard</a></b>
</div>
"""

##################################################
# Example Papers
##################################################
example_papers = [
    {
        "title": "Attention Is All You Need",
        "abstract": (
            "The dominant sequence transduction models are based on complex recurrent or "
            "convolutional neural networks that include an encoder and a decoder. The best performing "
            "models also connect the encoder and decoder through an attention mechanism. We propose a "
            "new simple network architecture, the Transformer, based solely on attention mechanisms, "
            "dispensing with recurrence and convolutions entirely. Experiments on two machine "
            "translation tasks show these models to be superior in quality while being more "
            "parallelizable and requiring significantly less time to train."
        ),
        "score": 0.982,
        "note": "Revolutionary paper that introduced the Transformer architecture."
    },
    {
        "title": "Language Models are Few-Shot Learners",
        "abstract": (
            "Recent work has demonstrated substantial gains on many NLP tasks and benchmarks by "
            "pre-training on a large corpus of text followed by fine-tuning on a specific task. While "
            "typically task-agnostic in architecture, this method still requires task-specific "
            "fine-tuning datasets of thousands or tens of thousands of examples. By contrast, humans "
            "can generally perform a new language task from only a few examples or from simple "
            "instructions—something which current NLP systems still largely struggle to do. Here we "
            "show that scaling up language models greatly improves task-agnostic, few-shot "
            "performance, sometimes even reaching competitiveness with prior state-of-the-art "
            "fine-tuning approaches."
        ),
        "score": 0.956,
        "note": "Groundbreaking GPT-3 paper on few-shot learning."
    },
    {
        "title": "An Empirical Study of Neural Network Training Protocols",
        "abstract": (
            "This paper presents a comparative analysis of different training protocols for neural "
            "networks across various architectures. We examine the effects of learning rate schedules, "
            "batch size selection, and optimization algorithms on model convergence and final "
            "performance. Our experiments span multiple datasets and model sizes, providing practical "
            "insights for deep learning practitioners."
        ),
        "score": 0.623,
        "note": "Solid empirical comparison of training protocols."
    }
]

##################################################
# Build the Gradio Interface
##################################################
with gr.Blocks(theme=gr.themes.Default(), css=css) as iface:
    # Add the social media links and leaderboard link at the top
    gr.HTML(header_html)
    
    gr.Markdown("<div class='main-title'>Papers Impact: AI-Powered Research Impact Predictor</div>")
    gr.Markdown("**Predict the potential research impact (0–1) from title & abstract.**")

    with gr.Row():
        with gr.Column(elem_classes="input-section"):
            gr.Markdown("### Import from arXiv")
            with gr.Group(elem_classes="arxiv-input"):
                arxiv_input = gr.Textbox(
                    lines=1,
                    placeholder="e.g. 2504.11651",
                    label="arXiv URL or ID",
                    value="2504.11651"
                )
                gr.Markdown(
                    """
                    <p>
                      Enter an arXiv ID or URL. For example: 
                      <code>2504.11651</code> or <code>https://arxiv.org/pdf/2504.11651</code>
                    </p>
                    """
                )
                fetch_btn = gr.Button("🔍 Fetch Paper Details", variant="secondary")

            gr.Markdown("### Or Enter Manually")
            title_input = gr.Textbox(
                lines=2,
                placeholder="Paper title (≥3 words)...",
                label="Paper Title"
            )
            abs_input = gr.Textbox(
                lines=5,
                placeholder="Paper abstract (≥50 words)...",
                label="Paper Abstract"
            )
            status_box = gr.Textbox(label="Validation Status", interactive=False)
            predict_btn = gr.Button("🎯 Predict Impact", interactive=False, variant="primary")

        with gr.Column(elem_classes="result-section"):
            score_box = gr.Number(label="Impact Score")
            grade_box = gr.Textbox(label="Grade", elem_classes="grade-display")

    ############## METHODOLOGY EXPLANATION ##############
    gr.Markdown(
        """
        ### Scientific Methodology
        - **Training Data**: Model trained on an extensive dataset of published papers in CS.CV, CS.CL, CS.AI
        - **Optimization**: NDCG optimization with Sigmoid activation and MSE loss
        - **Validation**: Cross-validated against historical citation data
        - **Architecture**: Advanced transformer-based (LLaMA derivative) textual encoder
        - **Metrics**: Quantitative analysis of citation patterns and research influence
        """
    )

    ############## RATING SCALE ##############
    gr.Markdown(
        """
        ### Rating Scale
        | Grade | Score Range | Description         | Emoji |
        |-------|-------------|---------------------|-------|
        | AAA   | 0.900–1.000 | **Exceptional**     | 🌟    |
        | AA    | 0.800–0.899 | **Very High**       | ⭐    |
        | A     | 0.650–0.799 | **High**            | ✨    |
        | BBB   | 0.600–0.649 | **Above Average**   | 🔵    |
        | BB    | 0.550–0.599 | **Moderate**        | 📘    |
        | B     | 0.500–0.549 | **Average**         | 📖    |
        | CCC   | 0.400–0.499 | **Below Average**   | 📝    |
        | CC    | 0.300–0.399 | **Low**             | ✏️    |
        | C     | <0.300      | **Limited**         | 📑    |
        """
    )

    ############## EXAMPLE PAPERS ##############
    gr.Markdown("### Example Papers")
    for paper in example_papers:
        gr.Markdown(
            f"**{paper['title']}**  \n"
            f"Score: {paper['score']} | Grade: {get_grade_and_emoji(paper['score'])}  \n"
            f"{paper['abstract']}  \n"
            f"*{paper['note']}*\n---"
        )

    ##################################################
    # Events
    ##################################################
    # Validation triggers
    title_input.change(update_button_status, [title_input, abs_input], [status_box, predict_btn])
    abs_input.change(update_button_status, [title_input, abs_input], [status_box, predict_btn])

    # arXiv fetch
    fetch_btn.click(process_arxiv_input, [arxiv_input], [title_input, abs_input, status_box])

    # Predict handler
    def run_predict(t, a):
        s = predict(t, a)
        return s, get_grade_and_emoji(s)

    predict_btn.click(run_predict, [title_input, abs_input], [score_box, grade_box])

##################################################
# Launch
##################################################
if __name__ == "__main__":
    iface.launch()