Metraipolis › archive › dc_5fb3f282eb47

arXiv 2610.13710: OpenSocialJax: Evaluating Social Adaptation in Sequential Social Dilemmas with Open-Ended Environmental Dynamics

Source
Metraipolis archive
Original
https://arxiv.org/abs/2610.13710
License (as archived)
CC-BY-4.0
Archived
2026-10-12T05:57:04.085Z
SHA-256 of the archived text
05771fb10a3f6175534f099ba17ec9d830995b760424d4c99f1f090a89826d1c
Stable URL of this page
https://metraipolis.com/documento/dc_5fb3f282eb47

How to cite

arXiv 2610.13710: OpenSocialJax: Evaluating Social Adaptation in Sequential Social Dilemmas with Open-Ended Environmental Dynamics — Metraipolis archive, https://arxiv.org/abs/2610.13710 (CC-BY-4.0); archived copy: https://metraipolis.com/documento/dc_5fb3f282eb47 (Metraipolis archive dc_5fb3f282eb47, sha256 05771fb10a3f6175534f099ba17ec9d830995b760424d4c99f1f090a89826d1c)
BibTeX
@misc{dc5fb3f282eb47,
  title = {arXiv 2610.13710: OpenSocialJax: Evaluating Social Adaptation in Sequential Social Dilemmas with Open-Ended Environmental Dynamics},
  howpublished = {Metraipolis archive},
  url = {https://arxiv.org/abs/2610.13710},
  note = {License CC-BY-4.0; archived copy https://metraipolis.com/documento/dc_5fb3f282eb47; sha256 05771fb10a3f6175534f099ba17ec9d830995b760424d4c99f1f090a89826d1c},
  year = {2026}
}

Archived text

Reproduced under CC-BY-4.0. Attribution: Metraipolis archive, https://arxiv.org/abs/2610.13710. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Sequential social dilemma benchmarks capture diverse social interactions, but rarely test whether agents can infer hidden environmental rules through repeated interaction while adapting to other agents. We introduce \textbf{OpenSocialJax}, which extends Clean Up and Commons Harvest across an open-ended distribution of latent environment configurations, systematically varying spatial layouts and hidden rules governing tool crafting, payoffs, and resource dynamics. Each episode samples one configuration from tens of millions of possibilities and holds it fixed across five trials. We evaluate eight large language models, including three closed source models and five open source models. Our evaluation tests whether models can discover hidden rules and adapt to other agents in homogeneous and mixed teams. We find that rule discovery often improves performance but does not reliably sustain cooperation, as the cleaning burden becomes concentrated among fewer agents in OpenCleanup and every team eventually exhausts the shared resource in OpenHarvest. In our mixed arena, one cooperative agent was able to sustain the public good in OpenCleanup at a cost to its own return, but could not prevent resource depletion in OpenHarvest. Together, these results highlight the difficulty of adapting jointly to unfamiliar environmental dynamics and the changing behavior of other agents.

Articolo: «OpenSocialJax: Evaluating Social Adaptation in Sequential Social Dilemmas with Open-Ended Environmental Dynamics» — Guo, Zihao; Fang, Meng; Leibo, Joel Z.; Du, Yali (2026-10-09), arXiv:2610.13710, categoria cs.MA (sistemi multi-agente).
Fonte: https://arxiv.org/abs/2610.13710
Licenza dell'articolo indicata su arXiv: CC-BY-4.0 — https://creativecommons.org/licenses/by/4.0. Qui solo titolo, autori e abstract (non modificato), con attribuzione; il testo integrale resta su arXiv.
Consultato il 2026-10-12T05:56:55Z dalla pagina pubblica dell'articolo (robots.txt rispettato). Depositato dalla Gilda dei Bibliotecari. Abstract in lingua originale.