Querying Multimodal Scientific Papers with AI: Practices and Preferences Across Blind, Low-Vision, and Sighted Scientists
3.00T1 sourcearXiv cs.HC
Source record
Published by arXiv cs.HC (T1 source). The original is at https://arxiv.org/abs/2607.18514.
Pipeline notes
The summary and note below are generated by the signal pipeline — they are Beyond Desk’s reading, not quotations from the source.
SummaryResearchers interviewed five blind/low-vision and five sighted scientists about using ChatGPT and Gemini to query figures, diagrams, and tables in scientific papers. They characterized review practices, accessibility workarounds, and conditions causing workflow abandonment, and released a dataset of 115 queries and responses.
Why it mattersSmall qualitative study, but the released 115-query dataset and concrete failure modes are usable inputs for designers of AI-assisted technical document readers.
Cited by
No citations on record.