{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:3H47IKR7HOFWCZJ3GZVMUDHYDQ","short_pith_number":"pith:3H47IKR7","schema_version":"1.0","canonical_sha256":"d9f9f42a3f3b8b61653b366aca0cf81c30678071a609651f07eda60dc1a642e0","source":{"kind":"arxiv","id":"2302.02083","version":7},"attestation_state":"computed","paper":{"title":"Evaluating Large Language Models in Theory of Mind Tasks","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CY","cs.HC"],"primary_cat":"cs.CL","authors_text":"Michal Kosinski","submitted_at":"2023-02-04T03:50:01Z","abstract_excerpt":"Eleven Large Language Models (LLMs) were assessed using a custom-made battery of false-belief tasks, considered a gold standard in testing Theory of Mind (ToM) in humans. The battery included 640 prompts spread across 40 diverse tasks, each one including a false-belief scenario, three closely matched true-belief control scenarios, and the reversed versions of all four. To solve a single task, a model needed to correctly answer 16 prompts across all eight scenarios. Smaller and older models solved no tasks; GPT-3-davinci-003 (from November 2022) and ChatGPT-3.5-turbo (from March 2023) solved 20"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2302.02083","kind":"arxiv","version":7},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2023-02-04T03:50:01Z","cross_cats_sorted":["cs.CY","cs.HC"],"title_canon_sha256":"a58b3e31f00b6fe0cb16a6afe4cae6bd32065f0826437f4e2998c433ef55ff05","abstract_canon_sha256":"2f6ae6f15b595f2433efe518dfcf559ed9d190ac5c965b8734ce4f6f8adc8fb0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:30:55.176255Z","signature_b64":"rJXiBuRMkKAlhdt5ETRnYE7ZOppfNuspu8S/4lTp/W3Gkm2A3F5q5w8E+dxsndYEN7JrBh5yBLWt39Xi1K5HAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d9f9f42a3f3b8b61653b366aca0cf81c30678071a609651f07eda60dc1a642e0","last_reissued_at":"2026-07-05T09:30:55.175845Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:30:55.175845Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Evaluating Large Language Models in Theory of Mind Tasks","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CY","cs.HC"],"primary_cat":"cs.CL","authors_text":"Michal Kosinski","submitted_at":"2023-02-04T03:50:01Z","abstract_excerpt":"Eleven Large Language Models (LLMs) were assessed using a custom-made battery of false-belief tasks, considered a gold standard in testing Theory of Mind (ToM) in humans. The battery included 640 prompts spread across 40 diverse tasks, each one including a false-belief scenario, three closely matched true-belief control scenarios, and the reversed versions of all four. To solve a single task, a model needed to correctly answer 16 prompts across all eight scenarios. Smaller and older models solved no tasks; GPT-3-davinci-003 (from November 2022) and ChatGPT-3.5-turbo (from March 2023) solved 20"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.02083","kind":"arxiv","version":7},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2302.02083/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2302.02083","created_at":"2026-07-05T09:30:55.175896+00:00"},{"alias_kind":"arxiv_version","alias_value":"2302.02083v7","created_at":"2026-07-05T09:30:55.175896+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.02083","created_at":"2026-07-05T09:30:55.175896+00:00"},{"alias_kind":"pith_short_12","alias_value":"3H47IKR7HOFW","created_at":"2026-07-05T09:30:55.175896+00:00"},{"alias_kind":"pith_short_16","alias_value":"3H47IKR7HOFWCZJ3","created_at":"2026-07-05T09:30:55.175896+00:00"},{"alias_kind":"pith_short_8","alias_value":"3H47IKR7","created_at":"2026-07-05T09:30:55.175896+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":29,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.01581","citing_title":"Beyond Skepticism: Evaluating LLMs Pedagogical Intent Reasoning with the Adaptive Pedagogical Vigilance Framework","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09092","citing_title":"From Shortcuts to Reasoning: Robust Post-Training of Theory of Mind with Reinforcement Learning","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05557","citing_title":"AURA: Intent-Directed Probing for Implicit-Need Surfacing in Situated LLM Agents","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2606.20603","citing_title":"A Survey of Large Language Models for Perception and Measurement of Human Psychology","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21739","citing_title":"AttuneBench: A Conversation-Based Benchmark for LLM Emotional Intelligence","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29495","citing_title":"Cognitive World Model for Progressive BDI/E Trajectory Evaluation of Conversational Agents","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2606.28524","citing_title":"Developmental Trajectories of Situation Modeling and Mentalizing in Transformer Language Models","ref_index":62,"is_internal_anchor":false},{"citing_arxiv_id":"2605.27593","citing_title":"Voluntary Collusion with Secret Tools in Competing LLM Agents","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2606.27981","citing_title":"ToxiREX: A Dataset on Toxic REasoning in ConteXt","ref_index":187,"is_internal_anchor":false},{"citing_arxiv_id":"2606.23840","citing_title":"Embodied Explainability and Ontological Obstacles: Why We Struggle to Explain the Answers of Large Language Models (LLMs)","ref_index":72,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22826","citing_title":"Evaluating Large Language Models in a Complex Hidden Role Game","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21739","citing_title":"AttuneBench: A Conversation-Based Benchmark for LLM Emotional Intelligence","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20423","citing_title":"OSCToM: RL-Guided Adversarial Generation for High-Order Theory of Mind","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09826","citing_title":"EnactToM: An Evolving Benchmark for Functional Theory of Mind in Embodied Agents","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19762","citing_title":"What Really Improves Mathematical Reasoning: Structured Reasoning Signals Beyond Pure Code","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2506.02387","citing_title":"VS-Bench: Evaluating VLMs for Strategic Abilities in Multi-Agent Environments","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2507.02935","citing_title":"Theory of Mind in Action: The Instruction Inference Task in Dynamic Human-Agent Collaboration","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2509.13324","citing_title":"Designing Psychometric Bias Measures for ChatBots: An Application to Racial Bias Measurement","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02334","citing_title":"Holos: A Web-Scale LLM-Based Multi-Agent System for the Agentic Web","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2309.16797","citing_title":"Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution","ref_index":184,"is_internal_anchor":false},{"citing_arxiv_id":"2401.10774","citing_title":"Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads","ref_index":106,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09826","citing_title":"EnactToM: An Evolving Benchmark for Functional Theory of Mind in Embodied Agents","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2604.23198","citing_title":"StoryTR: Narrative-Centric Video Temporal Retrieval with Theory of Mind Reasoning","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01329","citing_title":"Truth or Tribe: How In-group Favoritism Prioritize Facts in Persona Agents","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20582","citing_title":"Trust, Lies, and Long Memories: Emergent Social Dynamics and Reputation in Multi-Round Avalon with LLM Agents","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3H47IKR7HOFWCZJ3GZVMUDHYDQ","json":"https://pith.science/pith/3H47IKR7HOFWCZJ3GZVMUDHYDQ.json","graph_json":"https://pith.science/api/pith-number/3H47IKR7HOFWCZJ3GZVMUDHYDQ/graph.json","events_json":"https://pith.science/api/pith-number/3H47IKR7HOFWCZJ3GZVMUDHYDQ/events.json","paper":"https://pith.science/paper/3H47IKR7"},"agent_actions":{"view_html":"https://pith.science/pith/3H47IKR7HOFWCZJ3GZVMUDHYDQ","download_json":"https://pith.science/pith/3H47IKR7HOFWCZJ3GZVMUDHYDQ.json","view_paper":"https://pith.science/paper/3H47IKR7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2302.02083&json=true","fetch_graph":"https://pith.science/api/pith-number/3H47IKR7HOFWCZJ3GZVMUDHYDQ/graph.json","fetch_events":"https://pith.science/api/pith-number/3H47IKR7HOFWCZJ3GZVMUDHYDQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3H47IKR7HOFWCZJ3GZVMUDHYDQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3H47IKR7HOFWCZJ3GZVMUDHYDQ/action/storage_attestation","attest_author":"https://pith.science/pith/3H47IKR7HOFWCZJ3GZVMUDHYDQ/action/author_attestation","sign_citation":"https://pith.science/pith/3H47IKR7HOFWCZJ3GZVMUDHYDQ/action/citation_signature","submit_replication":"https://pith.science/pith/3H47IKR7HOFWCZJ3GZVMUDHYDQ/action/replication_record"}},"created_at":"2026-07-05T09:30:55.175896+00:00","updated_at":"2026-07-05T09:30:55.175896+00:00"}