Language Models Represent Beliefs of Self and Others

arXiv:2402.18496v1 Announce Type: cross
Abstract: Understanding and attributing mental states, known as Theory of Mind (ToM), emerges as a fundamental capability for human social reasoning. While Large Language Models (LLMs) appear to possess certain ToM abilities, the mechanisms underlying these capabilities remain elusive. In this study, we discover that it is possible to linearly decode the belief status from the perspectives of various agents through neural activations of language models, indicating the existence of internal representations of self and others' …

abstract arxiv capabilities capability cs.ai cs.cl human language language models large language large language models llms mind reasoning social study theory theory of mind tom type understanding