发表机构
Korea University（高丽大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究针对大语言模型心智理论推理中事件隐藏导致的性能下降问题，提出PICTURE提示方法，通过在思维链中明确角色知识缺失，使模型在错误信念任务上性能提升7.3%
AI 中文摘要
模拟类人心智理论（Theory of Mind, ToM）一直是自然语言处理（NLP）领域的长期难题。现有研究为解决该问题引入了事件隐藏（又称视角采择）的推理步骤，即问答前移除角色未知的事件。但因事件隐藏涉及严格的输出格式约束，会导致心智理论推理出现性能下降问题。为缓解该问题，本文提出生成无事件隐藏的自由形式视角采择解释，这带来了一个未被充分探索的挑战：大语言模型（LLMs）需抑制对角色未知事件的响应，因为无事件隐藏会使LLMs在整个推理过程中接触到这些事件。针对该挑战，本文假设并通过实验验证：若在推理过程中明确角色对事件的知识缺失，LLMs可实现此类抑制。基于该发现，本文提出PICTURE，一种能让LLMs在自由形式思维链（Chain-of-Thought, CoT）中生成角色知识缺失的新型提示方法。实验结果显示，在错误信念任务上，PICTURE较现有提示方法平均性能提升7.3%
英文摘要
Simulating human-like Theory of Mind (ToM) has been a longstanding problem in natural language processing (NLP). To address this, existing works introduce a reasoning step of event hiding (a.k.a. perspective-taking), where events unknown to a character are removed before question answering. However, resorting to event hiding for ToM reasoning presents a performance degradation issue due to the strict output format constraints involved in event hiding. To mitigate this issue, we propose generating perspective-taking outputs as free-form explanations without event hiding, but this poses a notable yet underexplored challenge: LLMs need to inhibit responses to events unknown to characters, because the absence of event hiding exposes LLMs to these events throughout reasoning. To address this challenge, we hypothesize and empirically verify that LLMs can achieve such inhibition if a character's lack of knowledge about events is made explicit during reasoning. Based on this finding, we introduce PICTURE, a new prompting method that enables LLMs to generate a character's lack of knowledge within free-form Chain-of-Thought (CoT). Experimental results show that PICTURE outperforms existing prompting methods by an average of 7.3% on false-belief tasks.
