Understanding How Paper Writers Use AI-Generated Captions in Figure Caption Writing
Ho Yin (Sam) Ng, Ting-Yao Hsu, Jiyoo Min, Sungchul Kim, Ryan A. Rossi, Tong Yu, Hyunggu Jung, Ting-Hao ‘Kenneth’ Huang · 2025
Read Understanding How Paper Writers Use AI-Generated Captions in Figure Caption Writing by Ho Yin (Sam) Ng, Ting-Yao Hsu, Jiyoo Min, Sungchul Kim, Ryan A. Rossi, Tong Yu, Hyunggu Jung, Ting-Hao ‘Kenneth’ Huang (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-06317.txt.
Description
Ng and a team including Huang study how paper authors use AI captions for figures in twenty twenty-five. Figures and captions are key parts of a scientific paper. Many published captions are weak, mainly due to lack of author attention. Prior AI work focused on readers judging captions, not on authors writing them.
Eighteen authors rewrote captions for two figures from their own recent papers. They could use captions made by state of the art AI models as a resource. Video records of the writing were analysed in detail. Authors often began by copying an AI caption and then refining it. They liked long, detail rich captions that blend text and visual content. Current models helped less on complex figures. The authors draw lessons for tools that support academic writing.
Catalog record
| CREATOR | Ho Yin (Sam) Ng, Ting-Yao Hsu, Jiyoo Min, Sungchul Kim, Ryan A. Rossi, Tong Yu, Hyunggu Jung, Ting-Hao ‘Kenneth’ Huang |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-06317