让语言模型在对抗中学会合作，揭示团队竞争能激发更强协作。
Super-additive Cooperation in Language Model Agents
- 设计虚拟竞赛，让语言模型团队在囚徒困境中反复互动。
- 团队内协作与外部竞争结合，使一次性合作率显著提升。
- 为多智能体AI系统设计提供新思路，适合研究社会行为的学者。
随着自主人工智能（AI）代理的前景日益临近，研究其合作倾向变得愈发重要。本研究受超加成合作理论启发，认为重复互动与群体间竞争共同促成了人类的合作行为。我们设计了一个虚拟锦标赛，将语言模型代理分组，在囚徒困境游戏中相互对战。通过模拟组内动态与外部竞争，发现这种组合显著提升了整体及初始的一次性合作水平（即单次互动中的合作倾向）。该研究为大型语言模型在复杂社会场景中制定策略和行动提供了新框架，并表明群体间竞争可能出人意料地促进合作行为。这些洞见对设计未来能有效协同并契合人类价值观的多智能体AI系统至关重要。源代码已公开于 https://github.com/pippot/Superadditive-cooperation-LLMs。
原文摘要 · Abstract (English)
With the prospect of autonomous artificial intelligence (AI) agents, studying their tendency for cooperative behavior becomes an increasingly relevant topic. This study is inspired by the super-additive cooperation theory, where the combined effects of repeated interactions and inter-group rivalry have been argued to be the cause for cooperative tendencies found in humans. We devised a virtual tournament where language model agents, grouped into teams, face each other in a Prisoner's Dilemma game. By simulating both internal team dynamics and external competition, we discovered that this blend substantially boosts both overall and initial, one-shot cooperation levels (the tendency to cooperate in one-off interactions). This research provides a novel framework for large language models to strategize and act in complex social scenarios and offers evidence for how intergroup competition can, counter-intuitively, result in more cooperative behavior. These insights are crucial for designing future multi-agent AI systems that can effectively work together and better align with human values. Source code is available at https://github.com/pippot/Superadditive-cooperation-LLMs.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。