Metraipolis › archive › dc_3bbae8229596

UMAS: System-Level Uncertainty Quantification for Multi-Agent LLM Systems — arXiv:2610.33013

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.33013
License (as archived): CC0-1.0
Archived: 2026-10-27T08:51:31.709Z
SHA-256 of the archived text: 898aa166013a3eb6456208c95de472da652fb0e9ccf0c8bedd84f4ec237a8be8
Stable URL of this page: https://metraipolis.com/documento/dc_3bbae8229596

How to cite

UMAS: System-Level Uncertainty Quantification for Multi-Agent LLM Systems — arXiv:2610.33013 — arXiv (en), https://arxiv.org/abs/2610.33013 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3bbae8229596 (Metraipolis archive dc_3bbae8229596, sha256 898aa166013a3eb6456208c95de472da652fb0e9ccf0c8bedd84f4ec237a8be8)

BibTeX

@misc{dc3bbae8229596,
  title = {UMAS: System-Level Uncertainty Quantification for Multi-Agent LLM Systems — arXiv:2610.33013},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.33013},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3bbae8229596; sha256 898aa166013a3eb6456208c95de472da652fb0e9ccf0c8bedd84f4ec237a8be8},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.33013. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Multi-Agent Systems (MAS) are a promising paradigm for improving the reasoning capabilities of large language models, yet they remain prone to hallucinations and erroneous outputs. Uncertainty Quantification (UQ) is crucial for assessing system reliability, but existing methods are largely designed for single-agent settings; how to exploit MAS structure for UQ while keeping inference cost-neutral remains underexplored. We propose UMAS, a parameter-free, cost-neutral, system-level UQ framework that estimates the credibility of inter-agent influence from each generation's intrinsic confidence. UMAS uses these credibility signals to modulate how uncertainty is updated at each node, and then aggregates node-level uncertainty into a system-level estimate. Extensive evaluations on Debate and DyLAN across diverse benchmarks show that UMAS consistently outperforms state-of-the-art baselines by an average of 10.54 AUROC points. Beyond uncertainty estimation, UMAS enables hallucination detection and uncertainty-aware answer selection, improving MAS accuracy by up to 12.6 points on specific tasks while enhancing reliability. Code is available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.MA, nuovi lavori del giorno).
Categorie arXiv: Multiagent Systems (cs.MA) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.33013 — 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 2026-10-27T08:51:13.278Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).