从独立到社会学习：刻画LLM的递归社会改进
原标题：From Solo to Social Learning: Characterizing Recursive Social Improvement in LLMs
AI 摘要
该论文提出「递归社会改进」概念，研究多个自利 LLM 智能体在各自追求自身奖励、共享同一 token 预算的情况下，能否通过互相复制和修订技能文件来提升整个群体的表现。在受控环境中，传统社会学习算法能从同伴获益，但测试的三个 LLM 反而比单独学习效率低 30–52%，探索过窄或 token 耗尽。让模型自行编写和修订技能后，观察同伴能改变改进方式，但同等成本下仍无法超越独立学习者，且技能交换使群体收敛到更少的独立发现。结论是 LLM 能通过复制同伴提高学习效率，但尚不能提高学习效果。
正文节选
From Solo to Social Learning: Characterizing Recursive Social Improvement in LLMs Abstract Large language models (LLMs) can now improve themselves by revising the instructions they follow, and LLM agents are increasingly orchestrated to work together on complex problems. However, self-improvement methods typically optimize one system at a time, and multi-agent frameworks often have every model work toward a shared goal. We ask a different question. When each agent pursues its own reward, can sel