A Systematic Review on the Evaluation of Large Language Models in Theory of Mind Tasks

TL;DR

This systematic review analyzes how large language models are evaluated on Theory of Mind tasks, highlighting current methodologies, challenges, and the emerging but incomplete capabilities of LLMs in mental state reasoning.

Contribution

It provides a comprehensive taxonomy of ToM evaluation benchmarks and critically assesses the limitations and progress of LLMs in replicating human-like mental state understanding.

Findings

LLMs show emerging competence in ToM tasks

Significant gaps remain in LLMs' ability to emulate human mental reasoning

Evaluation techniques vary widely and have inherent limitations

Abstract

In recent years, evaluating the Theory of Mind (ToM) capabilities of large language models (LLMs) has received significant attention within the research community. As the field rapidly evolves, navigating the diverse approaches and methodologies has become increasingly complex. This systematic review synthesizes current efforts to assess LLMs' ability to perform ToM tasks, an essential aspect of human cognition involving the attribution of mental states to oneself and others. Despite notable advancements, the proficiency of LLMs in ToM remains a contentious issue. By categorizing benchmarks and tasks through a taxonomy rooted in cognitive science, this review critically examines evaluation techniques, prompting strategies, and the inherent limitations of LLMs in replicating human-like mental state reasoning. A recurring theme in the literature reveals that while LLMs demonstrate emerging competence in ToM tasks, significant gaps persist in their emulation of human cognitive abilities.