The article by Matt-Heun Hong and Anamaria Crisan presents a study on Generative Large Language Models (LLMs) and their potential in data analysis. Specifically, the study focuses on the use of these models for creating and refining visualizations via conversational interfaces. The authors developed AI Threads, a multi-threaded analytic chatbot that enables analysts to manage conversational context and improve output efficacy. The usability of AI Threads was evaluated through a crowdsourced study and in-depth interviews with expert analysts. The findings highlight the potential of LLMs and present future research opportunities.
Publication date: 9 Nov 2023
Project Page: https://osf.io/6wxpa/
Paper: https://arxiv.org/pdf/2311.05590
