mlsanityData to 7 Oct 2026

A Survey of Theory of Mind in Large Language Models: Evaluations, Representations, and Safety Risks

arXiv 2502.06470 ↗First posted 10 Feb 2025cs.CL cs.AIcited by 8

Not on the map. Why?

Abstract (arXiv)

Theory of Mind (ToM), the ability to attribute mental states to others and predict their behaviour, is fundamental to social intelligence. In this paper, we survey studies evaluating behavioural and representational ToM in Large Language Models (LLMs), identify important safety risks from advanced LLM ToM capabilities, and suggest several research directions for effective evaluation and mitigation of these risks.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 24 of its 36 references that are in the map. First posted on arXiv 10 Feb 2025.

Builds on

24 of 36 references.

References · 24

  1. 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,616
  2. 2023Generative Agents: Interactive Simulacra of Human Behaviorcited by 3,500
  3. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameterscited by 1,928
  4. 2023Representation Engineering: A Top-Down Approach to AI Transparencycited by 1,239
  5. 2023Steering Language Models With Activation Engineeringcited by 824
  6. 2018AI safety via debatecited by 384
  7. 2023Large Language Models Fail on Trivial Alterations to Theory-of-Mind Taskscited by 288
  8. 2019Risks from Learned Optimization in Advanced Machine Learning Systemscited by 250
  9. 2024A Careful Examination of Large Language Model Performance on Grade School Arithmeticcited by 188
  10. 2023Understanding Social Reasoning in Language Models with Language Modelscited by 166
  11. 2023FANToM: A Benchmark for Stress-testing Machine Theory of Mind in Interactionscited by 162
  12. 2023Large Language Models can Strategically Deceive their Users when Put Under Pressurecited by 132
12 more
  1. 2024AI Sandbagging: Language Models can Strategically Underperform on Evaluationscited by 131
  2. 2024Secret Collusion among AI Agents: Multi-Agent Deception via Steganographycited by 107
  3. 2024On scalable oversight with weak LLMs judging strong LLMscited by 94
  4. 2024Coercing LLMs to do and reveal (almost) anythingcited by 88
  5. 2024Defending Against Unforeseen Failure Modes with Latent Adversarial Trainingcited by 82
  6. 2024Transformers represent belief state geometry in their residual streamcited by 73
  7. 2024Escalation Risks from Language Models in Military and Diplomatic Decision-Makingcited by 61
  8. 2024LLMs achieve adult human performance on higher-order theory of mind taskscited by 59
  9. 2024Uncovering Deceptive Tendencies in Language Models: A Simulated Company AI Assistantcited by 32
  10. 2023Intelligent Virtual Assistants with LLM-based Process Automationcited by 26
  11. 2023Unveiling Theory of Mind in Large Language Models: A Parallel to Single Neurons in the Human Braincited by 18
  12. 2024Brittle Minds, Fixable Activations: Understanding Belief Representations in Language Modelscited by 8

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 2; 2026: 6.

8 papers.

  1. 2025Agents Require Metacognitive and Strategic Reasoning to Succeed in the Coming Labor Marketscited by 2
  2. 2026To Think or Not To Think, That is The Question for Large Reasoning Models in Theory of Mind Taskscited by 1
  3. 2025OnlineMate: An LLM-Based Multi-Agent Companion System for Cognitive Support in Online Learningcited by 0
  4. 2026Are LLMs Smarter Than Chimpanzees? An Evaluation on Perspective Taking and Knowledge State Estimationcited by 0
  5. 2026Do Large Language Models Get Caught in Hofstadter-Mobius Loops?cited by 0
  6. 2026Does Theory of Mind Improvement Really Benefit Human-AI Interactions? Empirical Findings from Interactive Evaluationscited by 0
  7. 2026Toward Human-Centered Multi-Agent Systems: Integrating Cognition, Culture, Values, and Cooperation in AI Agentscited by 0
  8. 2026ARGUS: Theory-of-Mind Guided Argument Generation with Strategy-Aware Planning and Knowledge Groundingcited by 0