arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2310.02124cs.CLcs.AIcs.CYcs.LGcs.MA
收藏

探索LLM智能体的协作机制:社会心理学视角

Exploring Collaboration Mechanisms for LLM Agents: A Social Psychology View

Zhejiang University(浙江大学);Google DeepMind(谷歌DeepMind);National University of Singapore(新加坡国立大学)

Jintian Zhang, Xin Xu, Ningyu Zhang, Ruibo Liu, Bryan Hooi, Shumin Deng

AI总结:

本文从社会心理学视角,通过构建具有不同特质和思维模式的LLM智能体社会,在三个基准数据集上验证了协作策略能超越先进方法并提高效率,同时揭示智能体表现出类似人类的社会行为。

摘要:

随着自然语言处理(NLP)系统日益被应用于复杂的社会环境中,一个紧迫的问题浮现出来:这些NLP系统能否在由多个大型语言模型(LLM)组成的多智能体社会中,展现出类似人类的协作智能?本文通过将实践实验与理论见解相结合,探讨了当代NLP系统之间的协作机制。我们构建了四个由LLM智能体组成的独特“社会”,每个智能体具有特定的“特质”(随和或过度自信),并以不同的“思维模式”(辩论或反思)参与协作。通过在三个基准数据集上评估这些多智能体社会,我们发现某些协作策略不仅超越了以往最先进的方法,还优化了效率(使用更少的API令牌)。此外,我们的结果进一步表明,LLM智能体表现出类似人类的社会行为,如从众和达成共识,这反映了基础的社会心理学理论。最后,我们整合社会心理学的见解来情境化LLM智能体的协作,激发对LLM协作机制的进一步研究。我们承诺分享我们的代码和数据集(https://github.com/zjunlp/MachineSoM),希望促进这一有前景方向的研究。

英文摘要:

As Natural Language Processing (NLP) systems are increasingly employed in intricate social environments, a pressing query emerges: Can these NLP systems mirror human-esque collaborative intelligence, in a multi-agent society consisting of multiple large language models (LLMs)? This paper probes the collaboration mechanisms among contemporary NLP systems by melding practical experiments with theoretical insights. We fabricate four unique `societies' comprised of LLM agents, where each agent is characterized by a specific `trait' (easy-going or overconfident) and engages in collaboration with a distinct `thinking pattern' (debate or reflection). Through evaluating these multi-agent societies on three benchmark datasets, we discern that certain collaborative strategies not only outshine previous top-tier approaches, but also optimize efficiency (using fewer API tokens). Moreover, our results further illustrate that LLM agents manifest human-like social behaviors, such as conformity and consensus reaching, mirroring foundational social psychology theories. In conclusion, we integrate insights from social psychology to contextualize the collaboration of LLM agents, inspiring further investigations into the collaboration mechanism for LLMs. We commit to sharing our code and datasets\footnote{\url{https://github.com/zjunlp/MachineSoM}.}, hoping to catalyze further research in this promising avenue.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑