Arxiv Paper Summarizer

A web application to fetch and translate summaries from Arxiv.org papers using GPT-3.5.

Features

• Modern and responsive web interface
• Paper search by category
• Automatic translation of titles and summaries
• PDF preview
• Minimalist design with TailwindCSS

Requirements

• Python 3.13
• OpenAI API Key

Installation

1. Clone the repository
2. Create a virtual environment:

   python -m venv venv
   source venv/bin/activate  # On Unix/macOS

3. Install dependencies:

   pip install -r requirements.txt

4. Create .env file with your OpenAI API key:

   OPENAI_API_KEY=your_api_key_here
   

Usage

1. Start the server:

   python main.py

2. Open in browser: http://localhost:5000

Web Interface

The web interface allows you to:

• Select the paper category to search
• Specify the number of results
• View paper summaries with a clean and modern design
• Get translations with a single click
• Access paper PDFs directly

API Endpoints

You can also use the API directly:

1. GET /papers/

   • Gets recent papers from Arxiv
   • Parameters:
     • category (optional): Arxiv category (default: "cs.AI")
     • max_results (optional): Maximum number of results (default: 5)
   • Example: http://localhost:5000/papers/?category=cs.AI&max_results=3

2. GET /papers/<paper_id>/summary

   • Gets a translated summary of a specific paper
   • Parameters:
     • translate (optional): Whether to translate the title (default: true)
   • Example: http://localhost:5000/papers/2312.01047/summary

API Response Example

{
    "original_title": "Deep Learning in Computer Vision",
    "translated_title": "Deep Learning in Computer Vision",
    "original_summary": "This paper presents...",
    "translated_summary": "This paper presents..."
}