CogToM: A Comprehensive Theory of Mind Benchmark inspired by Human Cognition for Large Language Models

TL;DR

CogToM is a comprehensive, multi-paradigm benchmark designed to evaluate large language models' Theory of Mind capabilities, revealing significant performance gaps and cognitive divergences from humans.

Contribution

This paper introduces CogToM, a large-scale, theoretically grounded ToM benchmark with over 8000 bilingual instances across 46 paradigms, filling gaps in existing narrow assessments.

Findings

Significant performance heterogeneity among models

Persistent bottlenecks in specific cognitive dimensions

Potential divergences between LLM and human cognition

Abstract

Whether Large Language Models (LLMs) truly possess human-like Theory of Mind (ToM) capabilities has garnered increasing attention. However, existing benchmarks remain largely restricted to narrow paradigms like false belief tasks, failing to capture the full spectrum of human cognitive mechanisms. We introduce CogToM, a comprehensive, theoretically grounded benchmark comprising over 8000 bilingual instances across 46 paradigms, validated by 49 human annotator.A systematic evaluation of 22 representative models, including frontier models like GPT-5.1 and Qwen3-Max, reveals significant performance heterogeneities and highlights persistent bottlenecks in specific dimensions. Further analysis based on human cognitive patterns suggests potential divergences between LLM and human cognitive structures. CogToM offers a robust instrument and perspective for investigating the evolving cognitive boundaries of LLMs.