MeDxAgent : Multi-Agent Consultation for Interactive Medical Diagnosis
Akshat Sanghvi, Naren Akash, Raza Imam, Amit Sharma, Mohit Jain · 2026
Read MeDxAgent : Multi-Agent Consultation for Interactive Medical Diagnosis by Akshat Sanghvi, Naren Akash, Raza Imam, Amit Sharma, Mohit Jain (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-2606-03416.txt.
Description
Sanghvi and co-authors build a system that diagnoses illness through dialogue, not one-shot answers. Most tests of medical language models hand over full case details at once. Real clinic work is open-ended and step by step. Doctors refine guesses by asking targeted questions. The team aims to mirror that process with software.
They first build MeDxBench, a set of 4,421 clinical cases across 20 specialties. On it they test MeDxAgent, a multi-agent consultation system. Design choices at the prompt, flow, and agent levels are varied in turn. The system scores 10.3 percent higher than the baseline. It closes 52.3 percent of the gap to a full-information oracle. Asking for demographics first helps accuracy. Passing short dialogue summaries also helps, as does feeding candidate diagnoses into question choice. The gains appear in full only when the choices combine. Code and data are set for release on publication.
Catalog record
| CREATOR | Akshat Sanghvi, Naren Akash, Raza Imam, Amit Sharma, Mohit Jain |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 86 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2606.03416