Metraipolis › archive › dc_cd70f0ff9969

Believed but Not Said: Measuring Hidden Belief Collapse in Multi-Agent LLM Discussions — arXiv:2610.22138

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.22138
License (as archived): CC0-1.0
Archived: 2026-10-16T08:44:02.701Z
SHA-256 of the archived text: dcebd957d2f2a81eda052efdee13b75eeb55c39eb83d32570c676d363c0a6977
Stable URL of this page: https://metraipolis.com/documento/dc_cd70f0ff9969

How to cite

Believed but Not Said: Measuring Hidden Belief Collapse in Multi-Agent LLM Discussions — arXiv:2610.22138 — arXiv (en), https://arxiv.org/abs/2610.22138 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cd70f0ff9969 (Metraipolis archive dc_cd70f0ff9969, sha256 dcebd957d2f2a81eda052efdee13b75eeb55c39eb83d32570c676d363c0a6977)

BibTeX

@misc{dccd70f0ff9969,
  title = {Believed but Not Said: Measuring Hidden Belief Collapse in Multi-Agent LLM Discussions — arXiv:2610.22138},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.22138},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cd70f0ff9969; sha256 dcebd957d2f2a81eda052efdee13b75eeb55c39eb83d32570c676d363c0a6977},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.22138. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

LLM systems increasingly rely on groups of agents that discuss, critique, and check one another, assuming that peers will catch individual errors. We test this assumption against a minimal adversary. In a ten-agent factual discussion, one or three adversarial agents repeatedly assert a plausible false answer. After each round we probe each benign agent's private answer through a separate call whose response is hidden from its peers, and we separately label its public stance toward the claim. Across six models and three factual-QA datasets, groups remain stable without an attacker: only 0.6% of benign agents finish with the false answer. A single adversarial agent increases this rate to 44.8% and three to 60.5%. Studies of conformity typically report public agreement without private change; we find the reverse. With three adversarial agents present, benign agents hold the false answer in 51.6% of exposed turns but publicly endorse it in only 19.7%. Much of the influence therefore remains invisible in the discussion, and the false answers persist when agents are asked for their own views. Agents that retain the true answer often fail to defend it; once adopted, the false answer spreads among benign agents, and baseline knowledge does not predict robustness.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.MA, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL); Multiagent Systems (cs.MA).
Pagina del lavoro: https://arxiv.org/abs/2610.22138 — 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 2026-10-16T08:43:47.672Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).