arXiv cs.CL· Neemesh Yadav, Yihuai Lan, Shan Dong, Mai Hieu Hien, Palakorn Achananuparp, Jing Jiang, Ee-Peng Lim·· 13 小时前AI 评分29
心智理论与亲社会信念如何影响 LLM 在最后通牒博弈中的人类对齐行为
Effects of Theory of Mind and Prosocial Beliefs on Steering Human-Aligned Behaviors of LLMs in Ultimatum Games
AI 导读
研究用最后通牒博弈测试心智理论（ToM）推理能否让 LLM 智能体的行为更贴合人类规范，对 o3-mini、DeepSeek-R1 Distilled Qwen 32B 等多个模型做了 2,700 次模拟。
正文 · 原文
Abstract:Large Language Models (LLMs) have shown potential in simulating human behaviors and performing theory-of-mind (ToM) reasoning, crucial for complex social interactions. We investigate ToM reasoning's role in aligning agentic behaviors with human norms in negotiation tasks, using the ultimatum game as our referenced task. We initialized LLM agents with different prosocial beliefs (Greedy, Fair, Selfless) and reasoning methods (chain of thought and ToM reasoning of varying levels), examining their decision-making processes and outcomes across multiple LLMs, including reasoning models like o3-mini and DeepSeek-R1 Distilled Qwen 32B. We perform 2,700 simulations to show that ToM reasoning enhances behavioral alignment with human norms, decision-making consistency, and negotiation outcomes. Consistent with prior findings, reasoning LLMs exhibit limited capability compared to ToM-enhanced LLMs, with different game roles benefiting from different ToM orders. Agents almost always chose strategies consistent with their assigned beliefs, and doing so aligned better with human norms in nearly every role, yet Greedy and Selfless agents still fell short of their norms in outcomes. Human verification further revealed that Llama 3.3 70B produces reasoning most consistent with its actions and beliefs. Our findings advance understanding of ToM's role in human-AI interaction and cooperative decision-making. The code used for our experiments can be found at this https URL.
| Comments: | Accepted to AACL-IJCNLP 2026. v2 corrects the RQ4 analysis (Section 5.4, Fig. 4); other results remain unchanged |
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC) |
| Cite as: | arXiv:2505.24255 [cs.CL] |
| (or arXiv:2505.24255v3 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2505.24255 arXiv-issued DOI via DataCite |
Submission history
From: Palakorn Achananuparp [view email]
[v1]
Fri, 30 May 2025 06:23:52 UTC (3,380 KB)
[v2]
Mon, 24 Aug 2026 03:10:58 UTC (156 KB)
[v3]
Wed, 21 Oct 2026 12:23:50 UTC (155 KB)
来源：arXiv cs.CL · arxiv.org