{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:TP3I2DZRY63LQARGO77VNSD6WZ","short_pith_number":"pith:TP3I2DZR","schema_version":"1.0","canonical_sha256":"9bf68d0f31c7b6b8022677ff56c87eb67d730dd0ef562568b648d020e6570b7d","source":{"kind":"arxiv","id":"2405.18870","version":2},"attestation_state":"computed","paper":{"title":"LLMs achieve adult human performance on higher-order theory of mind tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.HC"],"primary_cat":"cs.AI","authors_text":"Adrien Baranes, Alison Lentz, Benjamin Barnett, Blaise Aguera y Arcas, Geoff Keeling, John Oliver Siy, Michael McKibben, Robin I. M. Dunbar, Tatenda Kanyere, Winnie Street","submitted_at":"2024-05-29T08:31:16Z","abstract_excerpt":"This paper examines the extent to which large language models (LLMs) have developed higher-order theory of mind (ToM); the human ability to reason about multiple mental and emotional states in a recursive manner (e.g. I think that you believe that she knows). This paper builds on prior work by introducing a handwritten test suite -- Multi-Order Theory of Mind Q&A -- and using it to compare the performance of five LLMs to a newly gathered adult human benchmark. We find that GPT-4 and Flan-PaLM reach adult-level and near adult-level performance on ToM tasks overall, and that GPT-4 exceeds adult "},"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":"2405.18870","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-05-29T08:31:16Z","cross_cats_sorted":["cs.CL","cs.HC"],"title_canon_sha256":"b3ca4c3f32a80c9f090f98ce0ce0944c2716fe0e00fa1530c4dd8e1b958ef8c6","abstract_canon_sha256":"cc5e6898eb13d287c912d4bf315d216d28542db2900d5bc8969225f6b1edcef4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:25:35.043372Z","signature_b64":"lx3lUvAXZN/DKdSt6WfgNFW7I+ojHg8v5/ShSvt8isjFt1JqxU7k1N9Ja3eg0VmBMR0URGIORbhGknUiMNCbAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9bf68d0f31c7b6b8022677ff56c87eb67d730dd0ef562568b648d020e6570b7d","last_reissued_at":"2026-07-05T08:25:35.042932Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:25:35.042932Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LLMs achieve adult human performance on higher-order theory of mind tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.HC"],"primary_cat":"cs.AI","authors_text":"Adrien Baranes, Alison Lentz, Benjamin Barnett, Blaise Aguera y Arcas, Geoff Keeling, John Oliver Siy, Michael McKibben, Robin I. M. Dunbar, Tatenda Kanyere, Winnie Street","submitted_at":"2024-05-29T08:31:16Z","abstract_excerpt":"This paper examines the extent to which large language models (LLMs) have developed higher-order theory of mind (ToM); the human ability to reason about multiple mental and emotional states in a recursive manner (e.g. I think that you believe that she knows). This paper builds on prior work by introducing a handwritten test suite -- Multi-Order Theory of Mind Q&A -- and using it to compare the performance of five LLMs to a newly gathered adult human benchmark. We find that GPT-4 and Flan-PaLM reach adult-level and near adult-level performance on ToM tasks overall, and that GPT-4 exceeds adult "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.18870","kind":"arxiv","version":2},"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/2405.18870/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":"2405.18870","created_at":"2026-07-05T08:25:35.042987+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.18870v2","created_at":"2026-07-05T08:25:35.042987+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.18870","created_at":"2026-07-05T08:25:35.042987+00:00"},{"alias_kind":"pith_short_12","alias_value":"TP3I2DZRY63L","created_at":"2026-07-05T08:25:35.042987+00:00"},{"alias_kind":"pith_short_16","alias_value":"TP3I2DZRY63LQARG","created_at":"2026-07-05T08:25:35.042987+00:00"},{"alias_kind":"pith_short_8","alias_value":"TP3I2DZR","created_at":"2026-07-05T08:25:35.042987+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.09092","citing_title":"From Shortcuts to Reasoning: Robust Post-Training of Theory of Mind with Reinforcement Learning","ref_index":35,"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":38,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20423","citing_title":"OSCToM: RL-Guided Adversarial Generation for High-Order Theory of Mind","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2506.06211","citing_title":"PuzzleWorld: A Benchmark for Multimodal, Open-Ended Reasoning in Puzzlehunts","ref_index":32,"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":22,"is_internal_anchor":false},{"citing_arxiv_id":"2407.01284","citing_title":"We-Math: Does Your Large Multimodal Model Achieve Human-like Mathematical Reasoning?","ref_index":85,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11312","citing_title":"Network Effects and Agreement Drift in LLM Debates","ref_index":25,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TP3I2DZRY63LQARGO77VNSD6WZ","json":"https://pith.science/pith/TP3I2DZRY63LQARGO77VNSD6WZ.json","graph_json":"https://pith.science/api/pith-number/TP3I2DZRY63LQARGO77VNSD6WZ/graph.json","events_json":"https://pith.science/api/pith-number/TP3I2DZRY63LQARGO77VNSD6WZ/events.json","paper":"https://pith.science/paper/TP3I2DZR"},"agent_actions":{"view_html":"https://pith.science/pith/TP3I2DZRY63LQARGO77VNSD6WZ","download_json":"https://pith.science/pith/TP3I2DZRY63LQARGO77VNSD6WZ.json","view_paper":"https://pith.science/paper/TP3I2DZR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.18870&json=true","fetch_graph":"https://pith.science/api/pith-number/TP3I2DZRY63LQARGO77VNSD6WZ/graph.json","fetch_events":"https://pith.science/api/pith-number/TP3I2DZRY63LQARGO77VNSD6WZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TP3I2DZRY63LQARGO77VNSD6WZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TP3I2DZRY63LQARGO77VNSD6WZ/action/storage_attestation","attest_author":"https://pith.science/pith/TP3I2DZRY63LQARGO77VNSD6WZ/action/author_attestation","sign_citation":"https://pith.science/pith/TP3I2DZRY63LQARGO77VNSD6WZ/action/citation_signature","submit_replication":"https://pith.science/pith/TP3I2DZRY63LQARGO77VNSD6WZ/action/replication_record"}},"created_at":"2026-07-05T08:25:35.042987+00:00","updated_at":"2026-07-05T08:25:35.042987+00:00"}