一项新的研究论文通过用大型语言模型（LLM）代理取代人类参与者来复制经济学实验，探讨了大型语言模型的市场行为。研究发现，与人类驱动的市场相比，由LLM代理组成市场向均衡收敛的速度较慢或根本不收敛，导致资源配置效率低下。对交易决策的分析揭示了不同模型家族和角色之间存在显著差异，在执行交易时，从战略调整转向紧迫性，这可以通过思维链（Chain-of-Thought）追踪来体现。 AI
影响 这项研究表明，在没有进一步的对齐研究之前，目前的LLM可能不适合直接部署到复杂的经济市场机制中。
排序理由 该集群包含一篇在arXiv上发表的详细介绍实验结果的研究论文。
在 arXiv cs.MA (Multiagent) 阅读 →
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
arXiv:2609.02580v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly deployed as economic agents, yet there is little evidence whether LLM agents are suited for participating in market mechanisms designed for humans, and whether these mechanisms deliver…
Large language models (LLMs) are increasingly deployed as economic agents, yet there is little evidence whether LLM agents are suited for participating in market mechanisms designed for humans, and whether these mechanisms deliver desired outcomes when faced with LLM agents. We a…
<p>Market mechanisms like double auctions rely on assumptions about participant behavior. Humans converge toward equilibrium prices through repeated rounds of bidding. A new paper from Struski et al. replaces human traders with LLM agents and finds those assumptions break. Market…