Location-Aware Pretraining for Medical Difference Visual Question Answering
Denis Musinguzi, Caren Han, Prasenjit Mitra · 2026
Read Location-Aware Pretraining for Medical Difference Visual Question Answering by Denis Musinguzi, Caren Han, Prasenjit Mitra (2026) 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-2603-04950.txt.
Description
Denis Musinguzi and the rest of the team present place-aware early training for medical gap visual question answering, where models compare images to spot change that matters in medicine. Standard side-by-side or sorting training leaves vision encoders missing subtle changes that tell true disease course apart from capture shifts. Their method adds auto referring phrases, grounded labels, and linked auto referring phrases, building fine-grained forms grounded in space. Joined with a language model, it leads medical gap VQA on chest X-ray images.
Catalog record
| CREATOR | Denis Musinguzi, Caren Han, Prasenjit Mitra |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33709 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.04950