Leveraging Large Language Models for Accessible Data Visualisation

Maliny Po on using large language models to automatically generate accessible text descriptions of statistical visualisations, at WOMBAT 2026.
Published

December 1, 2026

December 01, 03:00 PM

Leveraging Large Language Models for Accessible Data Visualisation

Data visualisations are widely used to explore and communicate data-driven insights. However, decoding and interpreting data visualisations relies on sight and visual perception, creating accessibility challenges for visually impaired individuals and screen reader users. This talk presents a work-in-progress framework that leverages OpenAI-based large language models and integrates visualisations, reference text, and source code to automatically generate accessible textual descriptions of statistical visualisations.

Register for the workshop day
Maliny Po

Maliny Po

Maliny Po is an Integrated PhD student in the Department of Econometrics and Business Statistics at Monash University. Her research interests include data visualisation, natural language processing (NLP), and large language models (LLM). She has contributed to projects involving data visualisation and LLMs, including developing a pipeline for bulk-processing legacy papers and generating alt text. Outside of academia, she enjoys reading and hiking.