---
title: AI Research Paper Chatbot
emoji: 📚
colorFrom: blue
colorTo: purple
sdk: gradio
sdk_version: 5.39.0
app_file: app.py
pinned: false
app_port: 7860
---

# 📚 AI Research Paper Chatbot

A modern conversational AI chatbot designed specifically for exploring and analyzing AI research papers. Features full paper text access, conversation memory, real-time streaming, and intelligent paper search.

## ✨ Latest Features

- 📖 **Smart Function Calling**: Intelligent paper retrieval using OpenAI's function calling API
- 🔍 **Dynamic Paper Fetching**: Automatically fetches full paper texts when needed
- 🧠 **Contextual Conversation Memory**: Maintains chat history with intelligent truncation
- 🚀 **Real-time Streaming**: Instant response streaming for better UX
- 🎛️ **Multiple Model Selection**: Choose between GPT-4o, GPT-4o-mini, and GPT-3.5 Turbo
- ⚙️ **Advanced Parameters**: Fine-tune temperature, max tokens, and top-p
- 🎨 **Modern UI**: Responsive design with intuitive controls
- 🛡️ **Robust Error Handling**: Clear error messages for common issues
- 📱 **Mobile Responsive**: Works great on all devices

## 🚀 Quick Start

### 1. Install Dependencies

```bash
pip install -r requirements.txt
```

### 2. Get OpenAI API Key

1. Visit [OpenAI Platform](https://platform.openai.com/api-keys)
2. Create an account or sign in
3. Generate a new API key
4. Copy the API key

### 3. Configure Environment

#### For Local Development
Set your OpenAI API key as an environment variable:

**Windows (PowerShell):**
```powershell
$env:OPENAI_API_KEY="your_openai_api_key_here"
```

**Windows (Command Prompt):**
```cmd
set OPENAI_API_KEY=your_openai_api_key_here
```

**Linux/macOS:**
```bash
export OPENAI_API_KEY="your_openai_api_key_here"
```

#### For Hugging Face Spaces Deployment
1. Go to your Space settings
2. Click on "Settings" tab
3. Scroll down to "Repository secrets"
4. Click "New secret"
5. **Name**: `OPENAI_API_KEY`
6. **Value**: Your actual OpenAI API key
7. Click "Add secret"

**Important**: Replace `your_openai_api_key_here` with your actual OpenAI API key.

### 4. Add Your Papers

Place your research paper text files in the `Papers/` directory. The system will automatically load all `.txt` files.

### 5. Run the Application

```bash
python app.py
```

The chatbot will be available at `http://localhost:7860`

## 🎯 Usage Guide

### Basic Paper Exploration
1. **Ask about specific topics**: "What papers discuss AI's impact on employment?"
2. **Request full papers**: "Show me the full paper about AI companions"
3. **Get detailed information**: "What's the conclusion of the pig disease detection paper?"
4. **Compare findings**: "Compare findings on AI in education"
5. **Ask for specific details**: "What methodology did they use in the pig disease paper?"

### Advanced Controls

#### Model Selection
- **GPT-4o-mini**: Fast, cost-effective (default)
- **GPT-4o**: Most capable, higher cost
- **GPT-3.5 Turbo**: Fastest, most affordable

#### Parameter Tuning
- **System Message**: Define AI personality and behavior
- **Max Tokens**: Control response length (1-4096)
- **Temperature**: Adjust creativity (0.0 = focused, 2.0 = creative)
- **Top-p**: Control response diversity (0.0-1.0)

#### Conversation Management
- **Clear Button**: Reset conversation history
- **Example Buttons**: Quick-start with sample messages

## 📚 Paper Database Features

### Automatic Paper Loading
- All `.txt` files in the `Papers/` directory are automatically loaded
- Paper titles are extracted from filenames
- Full text content is available for detailed analysis

### Intelligent Search
- **Keyword Matching**: Finds papers based on user query terms
- **Relevance Scoring**: Ranks papers by relevance to the query
- **Context-Aware**: Provides relevant paper excerpts for detailed responses

### Full Paper Access
- **Complete Text**: Access entire paper content when requested
- **Direct Quotes**: Get exact quotes from papers
- **Detailed Analysis**: Comprehensive answers including conclusions and methodology

## 🔧 Technical Details

### Latest OpenAI API Features
- **OpenAI SDK v1.98.0+**: Latest API patterns and features
- **Streaming Responses**: Real-time token streaming
- **Smart Retry Logic**: Automatic retry on failures
- **Timeout Handling**: 60-second request timeout
- **Error Classification**: Specific error messages for different issues

### Paper Processing
- **Automatic Loading**: Papers loaded at startup for fast access
- **Smart Search**: Keyword-based relevance scoring
- **Content Truncation**: Intelligent content selection for context
- **Full Text Access**: Complete paper retrieval when needed

### Conversation Memory
- **Intelligent Truncation**: Keeps recent messages while staying within limits
- **System Message Preservation**: Always maintains AI personality
- **Context Awareness**: Full conversation history for contextual responses

### Performance Optimizations
- **Async Processing**: Non-blocking UI during API calls
- **Memory Management**: Efficient conversation history handling
- **Error Recovery**: Graceful handling of API failures

## 🛠️ Configuration

### Environment Variables
```bash
OPENAI_API_KEY=your_api_key_here
```

### Model Parameters
```python
# Available models
AVAILABLE_MODELS = {
    "GPT-4o-mini": "gpt-4o-mini",
    "GPT-4o": "gpt-4o", 
    "GPT-3.5 Turbo": "gpt-3.5-turbo"
}
```

### Paper Directory Structure
```
Papers/
├── Paper Title 1.txt
├── Paper Title 2.txt
└── ...
```

## 🐛 Troubleshooting

### Common Issues

**API Key Errors**
- Ensure your `OPENAI_API_KEY` environment variable is set correctly
- Check that the API key has sufficient credits
- For Hugging Face Spaces: Verify the secret is named `OPENAI_API_KEY`

**Paper Loading Issues**
- Ensure papers are in `.txt` format
- Check that the `Papers/` directory exists
- Verify file encoding (UTF-8 recommended)

**Rate Limiting**
- Wait a moment and try again
- Consider using a different model

**Connection Issues**
- Check your internet connection
- Verify OpenAI API status at https://status.openai.com

**Memory Issues**
- Conversation history is maintained in memory during the session
- Long conversations are automatically truncated

### Error Messages
- **"Invalid API key"**: Check your environment variable or Hugging Face Spaces secrets
- **"Quota exceeded"**: Add credits to your OpenAI account
- **"Rate limit"**: Wait and retry
- **"Paper not found"**: Check that the paper file exists in the Papers directory

## 📊 Model Comparison

| Model | Speed | Cost | Capability | Best For |
|-------|-------|------|------------|----------|
| GPT-4o-mini | Fast | Low | Good | General chat, quick responses |
| GPT-4o | Medium | High | Excellent | Complex tasks, detailed analysis |
| GPT-3.5 Turbo | Fastest | Lowest | Good | Simple queries, high volume |

## 🔄 Recent Updates

- ✅ Added full paper text access functionality
- ✅ Implemented intelligent paper search
- ✅ Added automatic paper loading from Papers directory
- ✅ Enhanced system prompt with paper content
- ✅ Added example buttons for paper exploration
- ✅ Updated to OpenAI SDK v1.98.0+
- ✅ Added multiple model selection
- ✅ Improved error handling and messages
- ✅ Enhanced conversation memory management
- ✅ Added smart conversation truncation
- ✅ Modernized UI with better responsive design
- ✅ Fixed Pydantic compatibility issues
- ✅ Improved Hugging Face Spaces deployment

## 📝 License

This project is open source and available under the MIT License.