Monitoring and Steering Conversational LLM Agents in Data Analysis with On-the-Fly Code Visualization
Association for Computing Machinery (ACM) via YouTube
Overview
Watch a 13-minute conference talk from the 37th Annual ACM Symposium on User Interface Software and Technology (UIST 2024) exploring an innovative approach to monitoring and guiding conversational Large Language Model agents during data analysis processes. Learn about a novel system that enables real-time visualization of code execution while interacting with LLM agents, providing enhanced control and understanding of the analysis workflow. Discover how this technology allows users to better steer and supervise AI agents during complex data analysis tasks, bridging the gap between conversational AI interfaces and traditional programming environments. The presentation, delivered at UIST 2024 in Pittsburgh, PA, demonstrates practical applications and implications for the future of human-AI collaboration in data science.
Syllabus
WaitGPT: Monitoring and Steering Conversational LLM Agent in Data Analysis with On-the-Fly Code Visu
Taught by
ACM SIGCHI