MIRA: A Novel Framework for Fusing Modalities in Medical RAG
Jinhong Wang, Tajamul Ashraf, Zongyan Han, Jorma Laaksonen, Rao Mohammad Anwer · 2025
Read MIRA: A Novel Framework for Fusing Modalities in Medical RAG by Jinhong Wang, Tajamul Ashraf, Zongyan Han, Jorma Laaksonen, Rao Mohammad Anwer (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-2507-07902.txt.
Description
Jinhong Wang and co-authors fuse data types for medical question answering in this 2025 paper. Language models with vision can aid diagnosis, yet they often state facts that clash with medical knowledge. Fetching outside sources helps, but fetching too much or too little both hurt.
Their framework, MIRA, tunes how much is fetched and how it is fused with images and text. It guards against a subtler flaw: a model that first answers right, then talks itself into error by leaning on fetched text. The design aims for replies that stay true to established knowledge. Safer medical chatbots are the goal.
Catalog record
| CREATOR | Jinhong Wang, Tajamul Ashraf, Zongyan Han, Jorma Laaksonen, Rao Mohammad Anwer |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 58 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-07902