Metraipolis › archive › dc_fc471cef863e

Local Predictability and Collective Fidelity in LLM-Agent Societies — arXiv:2609.35813

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.35813
License (as archived)
CC0-1.0
Archived
2026-10-07T16:21:20.259Z
SHA-256 of the archived text
603f2c41ca6b5c396704005d43bc5e2e3f06d6905ab3e90cd0f0f8b59cab5379
Stable URL of this page
https://metraipolis.com/documento/dc_fc471cef863e

How to cite

Local Predictability and Collective Fidelity in LLM-Agent Societies — arXiv:2609.35813 — arXiv (en), https://arxiv.org/abs/2609.35813 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fc471cef863e (Metraipolis archive dc_fc471cef863e, sha256 603f2c41ca6b5c396704005d43bc5e2e3f06d6905ab3e90cd0f0f8b59cab5379)
BibTeX
@misc{dcfc471cef863e,
  title = {Local Predictability and Collective Fidelity in LLM-Agent Societies — arXiv:2609.35813},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.35813},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fc471cef863e; sha256 603f2c41ca6b5c396704005d43bc5e2e3f06d6905ab3e90cd0f0f8b59cab5379},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.35813. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Compact surrogates could reduce the cost of simulating large language model societies, but must reproduce collective behavior. We compare individual predictions and collective forecasts using 9,455 published trajectories and new experiments on opinion dynamics. Neighbor information improves individual prediction in all 16 public-data settings and pooled collective forecasts on held-out questions, although collective gains depend on transfer conditions. Tests on 24 new statements do not confirm earlier contrasting history effects in forecasts from the initial state. Qwen benefits from history after three observed rounds. These findings motivate direct collective validation, explicit limits on available observations, and comparisons with simple baselines.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.MA).
Categorie arXiv: Physics and Society (physics.soc-ph) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Multiagent Systems (cs.MA).
Pagina del lavoro: https://arxiv.org/abs/2609.35813 — 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 2026-10-07T15:54:11.812Z dalla pagina https://arxiv.org/catchup/cs.MA/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).