Metraipolis › archive › dc_5c6eb976b41b
Learning When to Collaborate: Selective Multi-Agent Medical Reasoning via Uncertainty-Aware Routing — arXiv:2610.33544
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.33544
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-27T07:56:39.380Z
- SHA-256 of the archived text
52b07fd182c7ef81684f930d2e83a8cbba21c3bed7cdf29b2d04559e30e6d6d4- Stable URL of this page
https://metraipolis.com/documento/dc_5c6eb976b41b
How to cite
Learning When to Collaborate: Selective Multi-Agent Medical Reasoning via Uncertainty-Aware Routing — arXiv:2610.33544 — arXiv (en), https://arxiv.org/abs/2610.33544 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5c6eb976b41b (Metraipolis archive dc_5c6eb976b41b, sha256 52b07fd182c7ef81684f930d2e83a8cbba21c3bed7cdf29b2d04559e30e6d6d4)
BibTeX
@misc{dc5c6eb976b41b,
title = {Learning When to Collaborate: Selective Multi-Agent Medical Reasoning via Uncertainty-Aware Routing — arXiv:2610.33544},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.33544},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5c6eb976b41b; sha256 52b07fd182c7ef81684f930d2e83a8cbba21c3bed7cdf29b2d04559e30e6d6d4},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.33544. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Multi-agent systems have emerged as a promising paradigm for multimodal medical reasoning, but existing methods rely on two assumptions that hinder practical deployment: (i) always-active dense collaboration among multiple agents, and (ii) dependence on large cloud-based models that raise privacy and efficiency concerns. In this paper, we challenge both assumptions and propose SMART-Med, a fully local multi-agent method that treats collaboration as a learnable decision rather than a default behavior. Our key insight is that not every medical query requires costly multi-agent deliberation: many can be reliably solved by a single well-trained agent, while only uncertain cases benefit from expert collaboration. Based on this insight, SMART-Med introduces an uncertainty-driven routing mechanism that adaptively decides when to answer directly and when to recruit a small subset of expert agents, effectively converting dense multi-agent reasoning into selective, query-adaptive coordination. To make this routing reliable, we curate a new multimodal medical dataset MedVR-3K with high-quality reasoning traces and a difficulty hierarchy, and then fine-tune local agents using GRPO with two complementary rewards: a certainty-aware accuracy reward and a vision-language reward model for correctness and reasoning quality respectively. Experiments on four medical visual question answering benchmarks show that SMART-Med achieves competitive performance compared to existing MAS approaches while substantially reducing inference cost (reducing inference time by more than 3.8x and token usage by 72%). Moreover, after fine-tuning on MedVR-3K, SMART-Med yields an average improvement of 6.7% across Slake, PathVQA, VQA-RAD, and PMC-VQA. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.33544 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-27T07:55:30.645Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).