mlsanityData to 13 Oct 2026

HI-TOM: A Benchmark for Evaluating Higher-Order Theory of Mind Reasoning in Large Language Models

arXiv 2310.16755 ↗First posted 25 Oct 2023cs.CL cs.AIcited by 61

On the map: LLM reasoning → Reasoning benchmarks and evaluation

Abstract (arXiv)

Theory of Mind (ToM) is the ability to reason about one's own and others' mental states. ToM plays a critical role in the development of intelligence, language understanding, and cognitive processes. While previous work has primarily focused on first and second-order ToM, we explore higher-order ToM, which involves recursive reasoning on others' beliefs. We introduce HI-TOM, a Higher Order Theory of Mind benchmark. Our experimental evaluation using various Large Language Models (LLMs) indicates a decline in performance on higher-order ToM tasks, demonstrating the limitations of current LLMs. We conduct a thorough analysis of different failure cases of LLMs, and share our thoughts on the implications of our findings on the future of NLP.

Key facts. Cited by 61 papers in the map (data through 13 Oct 2026). Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning. Builds on 21 of its 53 references that are in the map. First posted on arXiv 25 Oct 2023.

Builds on

21 of 53 references.

References · 21

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,304
  2. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,376
  3. 2022Large Language Models are Zero-Shot Reasonerscited by 4,348
  4. 2023Generative Agents: Interactive Simulacra of Human Behaviorcited by 3,541
  5. 2023QLoRA: Efficient Finetuning of Quantized LLMscited by 3,388
  6. 2023Sparks of Artificial General Intelligence: Early experiments with GPT-4cited by 2,586
  7. 2023Large Language Models are not Fair Evaluatorscited by 1,054
  8. 2023Multimodal Chain-of-Thought Reasoning in Language Modelscited by 746
  9. 2023Eliciting Latent Predictions from Transformers with the Tuned Lenscited by 600
  10. 2023Large Language Models Fail on Trivial Alterations to Theory-of-Mind Taskscited by 290
  11. 2022Data Contamination: From Memorization to Exploitationcited by 236
  12. 2022Neural Theory-of-Mind? On the Limits of Social Intelligence in Large LMscited by 220
9 more
  1. 2023Sources of Hallucination by Large Language Models on Inference Taskscited by 169
  2. 2023Can Large Language Models Infer Causation from Correlation?cited by 143
  3. 2021Thinking Fast and Slow: Efficient Text-to-Visual Retrieval with Transformerscited by 93
  4. 2018Evaluating Theory of Mind in Question Answeringcited by 64
  5. 2023Boosting Theory-of-Mind Performance in Large Language Models via Promptingcited by 64
  6. 2020Grounded Language Learning Fast and Slowcited by 54
  7. 2023Zero-shot Temporal Relation Extraction with ChatGPTcited by 52
  8. 2023Emergence of Maps in the Memories of Blind Navigation Agentscited by 31
  9. 2023Federated Prompting and Chain-of-Thought Reasoning for Improving LLMs Answeringcited by 14

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2024: 6; 2025: 32; 2026: 23.

61 papers.

Sort:

By area:

  1. 2024LLMs achieve adult human performance on higher-order theory of mind taskscited by 59
  2. 2025Embodied AI Agents: Modeling the Worldcited by 59
  3. 2024SimpleToM: Exposing the Gap between Explicit ToM Inference and Implicit ToM Application in LLMscited by 51
  4. 2025Language Models use Lookbacks to Track Beliefscited by 36
  5. 2025LongProc: Benchmarking Long-Context Language Models on Long Procedural Generationcited by 29
  6. 2025Hypothesis-Driven Theory-of-Mind Reasoning for Large Language Modelscited by 29
  7. 2024SPRIG: Improving Large Language Model Performance by System Prompt Optimizationcited by 28
  8. 2025AutoToM: Scaling Model-based Mental Inference via Automated Agent Modelingcited by 23
  9. 2025Re-evaluating Theory of Mind evaluation in large language modelscited by 23
  10. 2025EmoAgent: Assessing and Safeguarding Human-AI Interaction for Mental Health Safetycited by 21
  11. 2025Sentient Agent as a Judge: Evaluating Higher-Order Social Cognition in Large Language Modelscited by 21
  12. 2026Large Language Model Reasoning Failurescited by 21
38 more
  1. 2024LogicGame: Benchmarking Rule-Based Reasoning Abilities of Large Language Modelscited by 19
  2. 2025SoMi-ToM: Evaluating Multi-Perspective Theory of Mind in Embodied Social Interactionscited by 16
  3. 2024Position: Towards Bidirectional Human-AI Alignmentcited by 15
  4. 2025Social World Modelscited by 15
  5. 2025ToMAP: Training Opponent-Aware LLM Persuaders with Theory of Mindcited by 12
  6. 2025Large Language Models as Theory of Mind Aware Generative Agents with Counterfactual Reflectioncited by 11
  7. 2026OdysSim: Building Foundation Models for Human Behavior Simulationcited by 11
  8. 2025Persuasion Should be Double-Blind: A Multi-Domain Dialogue Dataset With Faithfulness Based on Causal Theory of Mindcited by 8
  9. 2025Do Theory of Mind Benchmarks Need Explicit Human-like Reasoning in Language Models?cited by 8
  10. 2025LLM-Hanabi: Evaluating Multi-Agent Gameplays with Theory-of-Mind and Rationale Inference in Imperfect Information Collaboration Gamecited by 8
  11. 2025AI Awarenesscited by 6
  12. 2025R^3-VQA: "Read the Room" by Video Social Reasoningcited by 6
  13. 2026AI4AI at Test-Time: Strong-to-Weak Capability Transfer via Harnessescited by 6
  14. 2024Belief in the Machine: Investigating Epistemological Blind Spots of Language Modelscited by 5
  15. 2026Social-R1: Towards Human-like Social Reasoning in LLMscited by 5
  16. 2025The Decrypto Benchmark for Multi-Agent Reasoning and Theory of Mindcited by 4
  17. 2025Investigating VLM Hallucination from a Cognitive Psychology Perspective: A First Step Toward Interpretation with Intriguing Observationscited by 4
  18. 2025MindPower: Enabling Theory-of-Mind Reasoning in VLM-based Embodied Agentscited by 4
  19. 2026Reinforcing Human Behavior Simulation via Verbal Feedbackcited by 4
  20. 2025Assessing LLMs in Art Contexts: Critique Generation and Theory of Mind Evaluationcited by 3
  21. 2025THiNK: Can Large Language Models Think-aloud?cited by 3
  22. 2026EnactToM: An Evolving Benchmark for Functional Theory of Mind in Embodied Agentscited by 3
  23. 2025Computer Vision Models Show Human-Like Sensitivity to Geometric and Topological Conceptscited by 2
  24. 2025HugAgent: A Human Simulation Benchmark for Individual-Level Reasoningcited by 2
  25. 2025Are Vision Language Models Cross-Cultural Theory of Mind Reasoners?cited by 1
  26. 2026DPBench: Structural Determinants of Multi-Agent LLM Coordination Under Simultaneous Resource Contentioncited by 1
  27. 2026OmniToM: Benchmarking Theory of Mind in LLMs via Explicit Belief Modelingcited by 1
  28. 2026A formal definition and meta-model for a machine theory of mindcited by 1
  29. 2026Inducing language models to assert their own consciousness restores human beliefs and valuescited by 1
  30. 2025Sensitivity Meets Sparsity: The Impact of Extremely Sparse Parameter Patterns on Theory-of-Mind of Large Language Modelscited by 0
  31. 2025textttDIAMONDs: A Dataset for mathbbDynamic mathbbInformation mathbbAnd mathbbMental modeling mathbbOf mathbbNumeric mathbbDiscussionscited by 0
  32. 2025TimeHC-RL: Temporal-aware Hierarchical Cognitive Reinforcement Learning for Enhancing LLMs' Social Intelligencecited by 0
  33. 2025Language Models Might Not Understand You: Evaluating Theory of Mind via Story Promptingcited by 0
  34. 2025Active Confusion Expression in Large Language Models: Leveraging World Models toward Better Social Reasoningcited by 0
  35. 2025Spot The Ball: A Benchmark for Visual Social Inferencecited by 0
  36. 2025S3IT: A Benchmark for Spatially Situated Social Intelligence Testcited by 0
  37. 2026CEI: A Benchmark for Evaluating Pragmatic Reasoning in Language Modelscited by 0
  38. 2026Dynamic Theory of Mind as a Temporal Memory Problem: Evidence from Large Language Modelscited by 0

50 most cited of 61 All 61 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2024ToM-LM: Delegating Theory of Mind Reasoning to External Symbolic Executors in Large Language Models
  2. 2026Understanding Artificial Theory of Mind: Perturbed Tasks and Reasoning in Large Language Models
  3. 2023MindGames: Targeting Theory of Mind in Large Language Models with Dynamic Epistemic Modal Logic
  4. 2024Mind Your Theory: Theory of Mind Goes Deeper Than Reasoning
  5. 2024Zero, Finite, and Infinite Belief History of Theory of Mind Reasoning in Large Language Models
  6. 2025Decompose-ToM: Enhancing Theory of Mind Reasoning in Large Language Models through Simulation and Task Decomposition
  7. 2023Understanding Social Reasoning in Language Models with Language Models
  8. 2024Multi-ToM: Evaluating Multilingual Theory of Mind Capabilities in Large Language Models