Adaptive Cluster Collaborativeness Boosts LLMs Medical Decision Support Capacity
Zhihao PengLiuxin BaoShengyuan Liu, Yixuan YuanChinese University of Hong Kong · 2025
Read Adaptive Cluster Collaborativeness Boosts LLMs Medical Decision Support Capacity by Zhihao PengLiuxin BaoShengyuan Liu, Yixuan YuanChinese University of Hong Kong (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-2507-21159.txt.
Description
Zhihao Peng and colleagues boost medical decision support with adaptive cluster collaboration among language models. Their 2025 arXiv paper builds multi-agent systems for healthcare. Groups of models can out-reason single models on hard cases. Static teamwork patterns, though, can fail and drag performance down. No standard pipeline existed for building the agent teams.
The Masked Agent Collaboration framework chooses agents by Pareto balance. Model size, speed, diversity, and throughput all enter that choice. Cross-consistency checks push agreement among cooperating agents. Collaborative information spreads in an adaptive, progressive way. The design aims to dodge the collapse of rigid teamwork. Medical choices gain from pooled, checked reasoning. Tests measure gains in decision support capacity.
Catalog record
| CREATOR | Zhihao PengLiuxin BaoShengyuan Liu, Yixuan YuanChinese University of Hong Kong |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 35 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-21159