TimeToM: Temporal Space is the Key to Unlocking the Door of Large Language Models' Theory-of-Mind

TL;DR

TimeToM introduces a temporal space framework with belief states to significantly enhance Large Language Models' ability to reason about mental states, addressing limitations of previous methods like Chain of Thought.

Contribution

The paper proposes a novel temporal space structure and a belief solver to improve LLMs' Theory of Mind reasoning, especially for higher-order beliefs.

Findings

Drastically improves ToM reasoning performance in LLMs.

Effectively handles higher-order ToM questions.

Advances towards coherent and robust ToM reasoning.

Abstract

Theory of Mind (ToM)-the cognitive ability to reason about mental states of ourselves and others, is the foundation of social interaction. Although ToM comes naturally to humans, it poses a significant challenge to even the most advanced Large Language Models (LLMs). Due to the complex logical chains in ToM reasoning, especially in higher-order ToM questions, simply utilizing reasoning methods like Chain of Thought (CoT) will not improve the ToM capabilities of LLMs. We present TimeToM, which constructs a temporal space and uses it as the foundation to improve the ToM capabilities of LLMs in multiple scenarios. Specifically, within the temporal space, we construct Temporal Belief State Chain (TBSC) for each character and inspired by the cognition perspective of the social world model, we divide TBSC into self-world beliefs and social world beliefs, aligning with first-order ToM (first-order beliefs) and higher-order ToM (higher-order beliefs) questions, respectively. Moreover, we design a novel tool-belief solver that, by considering belief communication between characters in temporal space, can transform a character's higher-order beliefs into another character's first-order beliefs under belief communication period. Experimental results indicate that TimeToM can dramatically improve the reasoning performance of LLMs on ToM questions while taking a big step towards coherent and robust ToM reasoning.