Metraipolis › archive › dc_94bcfdc41de7

Anticipating Social Consequences with Distributional World Models — arXiv:2610.30657

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.30657
License (as archived)
CC0-1.0
Archived
2026-10-23T15:12:56.961Z
SHA-256 of the archived text
943eaa2b193524922da3fbcabddd612fd6fd63e921b34590a7ffa52261b7dbd7
Stable URL of this page
https://metraipolis.com/documento/dc_94bcfdc41de7

How to cite

Anticipating Social Consequences with Distributional World Models — arXiv:2610.30657 — arXiv (en), https://arxiv.org/abs/2610.30657 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_94bcfdc41de7 (Metraipolis archive dc_94bcfdc41de7, sha256 943eaa2b193524922da3fbcabddd612fd6fd63e921b34590a7ffa52261b7dbd7)
BibTeX
@misc{dc94bcfdc41de7,
  title = {Anticipating Social Consequences with Distributional World Models — arXiv:2610.30657},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.30657},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_94bcfdc41de7; sha256 943eaa2b193524922da3fbcabddd612fd6fd63e921b34590a7ffa52261b7dbd7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.30657. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Social world models predict how interactions develop after an action, helping agents negotiate, cooperate and mediate. In these situations, one action can lead to several plausible futures, hence predicting only one future can overlook consequential alternatives, while exploring them through repeated simulations increases inference cost. We propose DSWM, a Distributional Social World Model that predicts alternative responses, their probabilities and immediate and final outcome values in one output per candidate action. We train a 4B model on repeated simulations using supervised fine-tuning and GRPO. At inference time, the actor uses these possible consequences to choose an action. Across three seen and two unseen environments, with interaction backbones absent from training collection, DSWM achieves the best average performance on 13 of 15 metrics. Ablations show that retaining multiple futures and their predicted probabilities improves outcomes in all five environments.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.30657 — 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 2026-10-23T15:09:08.381Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).