Metraipolis › archive › dc_d802b10f6f07
Do "True" and "False" Add Up to One? Complement Consistency in LLM-Agent Opinion Simulations — arXiv:2610.32874
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.32874
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-26T19:47:23.675Z
- SHA-256 of the archived text
380a0620476d6d68ff79b7cc012c894a591637986670647d4077f24573385b64- Stable URL of this page
https://metraipolis.com/documento/dc_d802b10f6f07
How to cite
Do "True" and "False" Add Up to One? Complement Consistency in LLM-Agent Opinion Simulations — arXiv:2610.32874 — arXiv (en), https://arxiv.org/abs/2610.32874 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d802b10f6f07 (Metraipolis archive dc_d802b10f6f07, sha256 380a0620476d6d68ff79b7cc012c894a591637986670647d4077f24573385b64)
BibTeX
@misc{dcd802b10f6f07,
title = {Do "True" and "False" Add Up to One? Complement Consistency in LLM-Agent Opinion Simulations — arXiv:2610.32874},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.32874},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d802b10f6f07; sha256 380a0620476d6d68ff79b7cc012c894a591637986670647d4077f24573385b64},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.32874. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
In LLM-agent opinion simulations, each agent reports the probability that a claim is true after seeing its own previous probability and the probabilities reported by its neighbors, and that number becomes its state in the next round. If the number describes the agent's position rather than the sentence it was shown, then rewriting the claim as its complement and replacing every probability p with 1 - p should return the complementary answer, so the two reports should sum to one. We test this with three small open models: Qwen3-4B, Llama-3.2-3B and Gemma-3-4B. On 2,304 agent states taken from a running network simulation, Qwen's paired reports sum to 1.18 on average, and on 48 real fact-checking claims from AVeriTeC they again exceed one. A controlled benchmark of 24 worlds, each with exactly two possible outcomes, shows the same failure in all three models. The discrepancy is two to ten times larger than the change produced by a same-meaning paraphrase, and it persists under repeated sampling and across peer group sizes and report strengths. When the reports are fed back into two populations that start from complementary states, the populations do not remain mirror images: their divergence is positive after five rounds for every model. Averaging each answer with the complement of its paired answer restores consistency by construction and lowers Brier loss averaged over the two wordings, but at the final round it beats the better of the two original wordings for only one of the three models. A schedule-by-targeting experiment on the same model gives a small, stable interaction because such sweeps keep the claim wording fixed. Consistency under rewording therefore needs its own test before simulated probabilities are read as beliefs 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); Social and Information Networks (cs.SI). Pagina del lavoro: https://arxiv.org/abs/2610.32874 — 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 2026-10-26T19:45:18.258Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).