Metraipolis › archive › dc_f9fcfeccf7ff

Does Multi-Agent Collaboration Reduce Hallucination in Clinical Tasks? — arXiv:2610.27027

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.27027
License (as archived): CC0-1.0
Archived: 2026-10-21T07:05:55.263Z
SHA-256 of the archived text: 3c2b26b3b9c2259c7ec03e41014356183eccd89e71d83d79ed96569e637b9034
Stable URL of this page: https://metraipolis.com/documento/dc_f9fcfeccf7ff

How to cite

Does Multi-Agent Collaboration Reduce Hallucination in Clinical Tasks? — arXiv:2610.27027 — arXiv (en), https://arxiv.org/abs/2610.27027 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f9fcfeccf7ff (Metraipolis archive dc_f9fcfeccf7ff, sha256 3c2b26b3b9c2259c7ec03e41014356183eccd89e71d83d79ed96569e637b9034)

BibTeX

@misc{dcf9fcfeccf7ff,
  title = {Does Multi-Agent Collaboration Reduce Hallucination in Clinical Tasks? — arXiv:2610.27027},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.27027},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f9fcfeccf7ff; sha256 3c2b26b3b9c2259c7ec03e41014356183eccd89e71d83d79ed96569e637b9034},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.27027. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Multi-agent LLM systems improve performance on clinical tasks, and that improvement is often read as evidence of a more trustworthy system. However, the two are not the same claim. A clinician acts on the stated reasoning and not only on the conclusion, so whether collaboration reduces hallucination, and what decides when it does, has to be measured in the reasoning too. Yet most evaluations score only the final answer. We evaluate four systems (a single-LLM baseline and three multi-agent topologies) across four models and three clinical tasks, scoring hallucination, under a factuality and faithfulness taxonomy, inside the rationale each system generates and reporting it alongside the task metric. We report that collaboration buys task performance but not lower hallucination as a rule, and no topology is reliably better than another. The effect differs by model: of the $36$ model--system--task comparisons, $12$ are large enough to resolve --- four decreases and eight increases, with no model on both sides. What sets the level, and the direction of any change, is the model rather than the topology or the number of agents. A self-correction experiment is consistent with this: shown its own flagged errors, every model except Llama-3.2-3B-Instruct reduces hallucination; the one model for which collaboration lowers hallucination gains the most, and Llama-3.2-3B-Instruct, the model whose hallucination it amplifies most, does not reduce its hallucination at all. We find that collaboration redistributes what its models can already do; it does not supply what they cannot.

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/2610.27027 — 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 2026-10-21T07:05:43.374Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).