arXiv cs.MA· Zihao Guo, Meng Fang, Joel Z. Leibo, Yali Du·· 6 天前AI 评分22
OpenSocialJax：在开放式环境动态的序贯社会困境中评估社会适应能力
OpenSocialJax: Evaluating Social Adaptation in Sequential Social Dilemmas with Open-Ended Environmental Dynamics
AI 导读
OpenSocialJax 将 Clean Up 和 Commons Harvest 扩展到包含数千万种潜在环境配置的开放式分布，每局从数千万种可能中采样一种配置并在五次试验中保持不变，系统改变空间布局及工具制作、收益和资源动态等隐藏规则。
正文 · 原文
Abstract: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.
| Comments: | 24 pages, including appendix |
| Subjects: | Multiagent Systems (cs.MA) |
| Cite as: | arXiv:2610.13710 [cs.MA] |
| (or arXiv:2610.13710v1 [cs.MA] for this version) | |
| https://doi.org/10.48550/arXiv.2610.13710 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Zihao Guo [view email]
[v1]
Fri, 9 Oct 2026 15:04:28 UTC (1,433 KB)
来源：arXiv cs.MA · arxiv.org