Metraipolis › archive › dc_77916b9bfb3d
Benchmarking large language model agent societies against human behavioural distributions — arXiv:2608.28182
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.28182
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T10:56:44.501Z
- SHA-256 of the archived text
234172f42b80dc487ceb55e206eb55eb23a6b911794d6ca3f46d3fa2f49fc2f5- Stable URL of this page
https://metraipolis.com/documento/dc_77916b9bfb3d
How to cite
Benchmarking large language model agent societies against human behavioural distributions — arXiv:2608.28182 — arXiv (en), https://arxiv.org/abs/2608.28182 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_77916b9bfb3d (Metraipolis archive dc_77916b9bfb3d, sha256 234172f42b80dc487ceb55e206eb55eb23a6b911794d6ca3f46d3fa2f49fc2f5)
BibTeX
@misc{dc77916b9bfb3d,
title = {Benchmarking large language model agent societies against human behavioural distributions — arXiv:2608.28182},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.28182},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_77916b9bfb3d; sha256 234172f42b80dc487ceb55e206eb55eb23a6b911794d6ca3f46d3fa2f49fc2f5},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.28182. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Populations of large language model agents are increasingly used as experimental societies. Three doubts shadow every such result: whether the agents behave like the humans they stand in for, whether a finding survives changes to the apparatus that leave the rules untouched, and whether apparent social dynamics are interaction at all rather than the reproduction of experiments the models have read. This article introduces SILICA, an open instrument that tests all three. Five environments carry published human anchors, each paired with perturbations that re-render the same rules and with variants whose payoffs point away from the memorised result. Twelve open-weight models were run through it on a single consumer graphics card. Agreement with human data is confined to starting points: first-round public-goods contributions fall inside the equivalence margin for eight of eleven models, while no model matches end-state contributions or the human corridor of cooperation. Merely swapping the order in which two actions are listed costs one model 58 points of cooperation. Presenting responders with a fixed schedule of offers shows that only one model, the sole reasoning-trained one, places its acceptance threshold where the incentive requires; two move theirs part of the way, two move them the wrong way, and three never acquire one. Conventions form through a shared prior over the names rather than through negotiation, though negotiation reappears once that prior is disrupted. On the certification ladder defined here, current silicon societies support exploratory claims and no more. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-31 in cs.CL). Categorie arXiv: Physics and Society (physics.soc-ph) ; Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2608.28182 — 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 2026-10-17T10:47:33.163Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-31?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).