AskTable: LLM-Powered Interactive Visual Data Story Construction from Tables

  • Gerile Aodeng
  • , Chenwei Liang
  • , Shaokun Zhang
  • , Yunshan Feng
  • , Qiyang Chen
  • , Guozheng Li*
  • , Chi Harold Liu
  • *Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

We present AskTable, an interactive system designed to support iterative construction of visual data stories for tabular data. AskTable enables users to explore, organize, and narrate data insights by combining structured subspace filtering, semantic insight graphs, and LLM-driven recommendations. The AskTable system employs a radial insight tree that visualizes user-driven analytical paths through a force-directed layout, allowing branching, modification, and backtracking during exploration. Users can interactively pin, query, and refine insights, while the system responds with relevant follow-up recommendations powered by a reasoning module, enabling the incremental construction of narrative data stories.

Original languageEnglish
Title of host publicationUIST Adjunct 2025 - Proceedings of the 38th Annual ACM Symposium on User Interface Software and Technology
EditorsAndrea Bianchi, Elena Glassman, Shengdong Zhao, Jeeeun Kim, Ian Oakley, Wendy E. Mackay
PublisherAssociation for Computing Machinery, Inc
ISBN (Electronic)9798400720369
DOIs
Publication statusPublished - 27 Sept 2025
Event38th Annual ACM Symposium on User Interface Software and Technology, UIST 2025 - Busan, Korea, Republic of
Duration: 28 Sept 20251 Oct 2025

Publication series

NameUIST Adjunct 2025 - Proceedings of the 38th Annual ACM Symposium on User Interface Software and Technology

Conference

Conference38th Annual ACM Symposium on User Interface Software and Technology, UIST 2025
Country/TerritoryKorea, Republic of
CityBusan
Period28/09/251/10/25

Keywords

  • exploratory data analysis
  • large language models
  • Tabular data
  • visual data story

Fingerprint

Dive into the research topics of 'AskTable: LLM-Powered Interactive Visual Data Story Construction from Tables'. Together they form a unique fingerprint.

Cite this