ConfReady: A RAG based Assistant and Dataset for Conference Checklist Responses

Michael Galarnyk*, Rutwik Routu*, Vidhyakshaya Kannan*, Kosha Bheda,
Prasun Banerjee, Agam Shah, Sudheer Chava
Georgia Institute of Technology

*Authors contributed equally

Overview

This repository:

• is an easy-to-use Llama or GPT powered web interface which can be used to empower authors to reflect on their work and assist authors with conference checklists
• is highly flexible and offers various adaptations and possibilities such as prompt customization, enabling developers to continue developing this tool for additional conferences

Installation

You can install ConfReady in two ways.

Prerequisites

• Python 3.11+
• NPM (Node.js. This is required for the web interface.

PyPI (Recommended)

pip install confready

GitHub Installation Steps

1. Clone the repository and navigate to the project directory:

   git clone https://github.com/gtfintechlab/ConfReady.git
   cd ConfReady

2. Run the installation script:

   source install.sh

3. Add your API keys:

   • Add LLM inference provider API keys to the .env file inside the server directory.

   TOGETHERAI_API_KEY=your_together_ai_key_here
   OPENAI_API_KEY=your_openai_key_here

Run the App

1. Run the Flask server:

   cd server
   python app.py

2. Run the Web Interface:

   cd confready
   npm start

3. Access the API:

   The server will be running on http://localhost:3000/.

Documentation

We welcome contributions to this library and encourage potential contributors to start by reviewing our documentation to familiarize themselves with the codebase's format and structure. Once you have a solid understanding, feel free to submit a pull request. We’re excited to collaborate with new contributors and drive this project forward together.

Citation

ConfReady was accepted at EMNLP Demonstrations 2025.

@misc{galarnyk2025confreadyragbasedassistant,
      title={ConfReady: A RAG based Assistant and Dataset for Conference Checklist Responses}, 
      author={Michael Galarnyk and Rutwik Routu and Vidhyakshaya Kannan and Kosha Bheda and Prasun Banerjee and Agam Shah and Sudheer Chava},
      year={2025},
      eprint={2408.04675},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2408.04675}, 
}