Arxiv Research Agent with Memory

This Streamlit app is a research assistant that helps you search for academic papers on arXiv. It keeps a memory of your past searches and the papers you've found, making it easy to keep track of your research.

The app uses a clean, tabbed interface for a smooth user experience. It leverages OpenAI's GPT-4o-mini model to format search results, the arxiv library for paper searching, and mem0 with a Qdrant vector database for persistent memory.

Features

• Tabbed Interface: A clean UI with dedicated "🔍 Search" and "🧠 Memory" tabs.
• arXiv Paper Search: Search for academic papers directly from the app.
• AI-Powered Formatting: Uses OpenAI's GPT-4o-mini to format search results into a clean, readable list.
• Persistent Search History: Remembers your past searches and the papers you found. Each search is saved as a distinct, formatted entry in the "Memory" tab, with the most recent searches appearing first.
• Persistent Configuration: Saves your OpenAI API key and username in a config.json file, so you don't have to enter them every time.
• Raw Memory Storage: Stores search history as pre-formatted markdown in a mem0 vectordb with Qdrant, ensuring consistent display.

How to Get Started

1. Clone the Repository

git clone https://github.com/Partha-SUST16/arxiv_agent_memory.git
cd arxiv_agent_memory

2. Install Dependencies

Make sure you have Python 3.8+ installed. Then, install the required packages:

pip install -r requirements.txt

3. Run Qdrant with Docker

The application uses Qdrant as its vector database. Make sure you have Docker installed and running.

Pull the Qdrant image:

docker pull qdrant/qdrant

Run the Qdrant container. This command also mounts a local directory (qdrant_storage) to persist the data.

docker run -p 6333:6333 -p 6334:6334 \
    -v $(pwd)/qdrant_storage:/qdrant/storage:z \
    qdrant/qdrant

Note: The app expects Qdrant on localhost:6333. If your setup differs, you can adjust the settings in config.json after the first run.

4. Run the Streamlit App

Launch the app with Streamlit:

streamlit run arxiv_agent.py

5. First-time Setup

The first time you run the app, it will create a config.json file.

1. Open the app in your browser.
2. Enter your OpenAI API Key in the input field.
3. Enter a Username.
4. These details will be saved to config.json for all future sessions.

Now you're all set to start your research!