Graphologue transforms LLM text responses into interactive, real-time diagrams, addressing the limitations of linear, text-based interfaces for complex information tasks. It targets users who need to quickly comprehend and flexibly interact with LLM-generated information, enabling non-linear dialogues for enhanced exploration and organization.

How It Works

Graphologue employs novel prompting strategies to extract entities and relationships from LLM responses, converting them into node-link diagrams. This approach facilitates graphical, non-linear dialogues, allowing users to interact with the diagrams to adjust presentation and submit context-specific prompts for further information.

Quick Start & Requirements

• Install dependencies: npm install
• Create .env file in the root with REACT_APP_OPENAI_API_KEY=sk-...
• Start local server: npm start
• Requires an OpenAI API key.
• Live Demo: [Link to Demo]
• UIST 2023 Paper: [Link to Paper]

Highlighted Details

• Real-time diagram generation from LLM responses.
• Interactive diagrams for flexible presentation and context-specific prompting.
• Facilitates non-linear dialogues for improved information exploration and comprehension.
• Based on research presented at UIST 2023.

Maintenance & Community

• Developed by researchers from UC San Diego's Creativity Lab.
• Supported by NSF grant #2009003.

Licensing & Compatibility

• License details are not explicitly stated in the README.
• Compatibility for commercial or closed-source use is not specified.

Limitations & Caveats

The project requires an OpenAI API key, and the README does not specify licensing terms, which may impact commercial adoption or integration into closed-source projects.