{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:ABTUNJ656YVCHXA6KQA37XHBAI","short_pith_number":"pith:ABTUNJ65","schema_version":"1.0","canonical_sha256":"006746a7ddf62a23dc1e5401bfdce1021cdb06d0ddb680ffcc9b41fdce1567cf","source":{"kind":"arxiv","id":"2302.08399","version":5},"attestation_state":"computed","paper":{"title":"Large Language Models Fail on Trivial Alterations to Theory-of-Mind Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Tomer Ullman","submitted_at":"2023-02-16T16:18:03Z","abstract_excerpt":"Intuitive psychology is a pillar of common-sense reasoning. The replication of this reasoning in machine intelligence is an important stepping-stone on the way to human-like artificial intelligence. Several recent tasks and benchmarks for examining this reasoning in Large-Large Models have focused in particular on belief attribution in Theory-of-Mind tasks. These tasks have shown both successes and failures. We consider in particular a recent purported success case, and show that small variations that maintain the principles of ToM turn the results on their head. We argue that in general, the "},"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":"2302.08399","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-02-16T16:18:03Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"8df100e0cbfe52dbf9f9ab25edbd7a77746aa831f7b4e45d15df76ec2652a289","abstract_canon_sha256":"af7d193173c53f7a0ccbd20129019b782a46527838dada5da11df475406b3d0a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:51:02.365685Z","signature_b64":"D8sdydLLyzSszYHe575jcVnU0VoiS+3Jf40QU8V7UbhbOwlQYiAQ78XCqeVaecbvfrVD9aFeB5jNDOsAyFKTDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"006746a7ddf62a23dc1e5401bfdce1021cdb06d0ddb680ffcc9b41fdce1567cf","last_reissued_at":"2026-07-05T05:51:02.365285Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:51:02.365285Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Large Language Models Fail on Trivial Alterations to Theory-of-Mind Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Tomer Ullman","submitted_at":"2023-02-16T16:18:03Z","abstract_excerpt":"Intuitive psychology is a pillar of common-sense reasoning. The replication of this reasoning in machine intelligence is an important stepping-stone on the way to human-like artificial intelligence. Several recent tasks and benchmarks for examining this reasoning in Large-Large Models have focused in particular on belief attribution in Theory-of-Mind tasks. These tasks have shown both successes and failures. We consider in particular a recent purported success case, and show that small variations that maintain the principles of ToM turn the results on their head. We argue that in general, the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.08399","kind":"arxiv","version":5},"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/2302.08399/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":"2302.08399","created_at":"2026-07-05T05:51:02.365353+00:00"},{"alias_kind":"arxiv_version","alias_value":"2302.08399v5","created_at":"2026-07-05T05:51:02.365353+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.08399","created_at":"2026-07-05T05:51:02.365353+00:00"},{"alias_kind":"pith_short_12","alias_value":"ABTUNJ656YVC","created_at":"2026-07-05T05:51:02.365353+00:00"},{"alias_kind":"pith_short_16","alias_value":"ABTUNJ656YVCHXA6","created_at":"2026-07-05T05:51:02.365353+00:00"},{"alias_kind":"pith_short_8","alias_value":"ABTUNJ65","created_at":"2026-07-05T05:51:02.365353+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":33,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.23339","citing_title":"When Robots Rate Their Own Interactions: Engagement Validity and the Strangeness Failure","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2606.13607","citing_title":"Reasoning as Pattern Matching: Shared Mechanisms in Human and LLM Everyday Reasoning","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2606.12721","citing_title":"The Theory of Mind Utility: Formal Specification of a Mentalizing Mechanism","ref_index":47,"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":32,"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":55,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00240","citing_title":"MindZero: Learning Online Mental Reasoning With Zero Annotations","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2606.31916","citing_title":"Theory of Mind and Persuasion Beyond Conversation: Assessing the Capacity of LLMs to Induce Belief States via Planning and Action","ref_index":31,"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":10,"is_internal_anchor":false},{"citing_arxiv_id":"2606.28524","citing_title":"Developmental Trajectories of Situation Modeling and Mentalizing in Transformer Language Models","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2606.27909","citing_title":"Triadic Werewolf: A Jester Role for Multi-Hop Theory of Mind in LLMs","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.27593","citing_title":"Voluntary Collusion with Secret Tools in Competing LLM Agents","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29512","citing_title":"MINDGAMES: A Live Arena for Evaluating Social and Strategic Reasoning in Multi-Agent LLMs","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30219","citing_title":"When Should Models Change Their Minds? Contextual Belief Management in Large Language Models","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23238","citing_title":"GENSTRAT: Toward a Science of Strategic Reasoning in Large Language Models","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2504.13898","citing_title":"Social Human Robot Embodied Conversation (SHREC) Dataset: Benchmarking Foundational Models' Social Reasoning","ref_index":54,"is_internal_anchor":false},{"citing_arxiv_id":"2506.18852","citing_title":"Mechanistic Interpretability Needs Philosophy","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20423","citing_title":"OSCToM: RL-Guided Adversarial Generation for High-Order Theory of Mind","ref_index":11,"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":31,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12920","citing_title":"Embodied Multi-Agent Coordination by Aligning World Models Through Dialogue","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18194","citing_title":"Beyond the Cartesian Illusion: Testing Two-Stage Multi-Modal Theory of Mind under Perceptual Bottlenecks","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17510","citing_title":"Scale-Dependent Collective Adaptation in Self-Amending LLM Societies: A Cross-Family Study of Emergent Governance","ref_index":63,"is_internal_anchor":false},{"citing_arxiv_id":"2506.17788","citing_title":"Bayesian Social Deduction with Graph-Informed Language Models","ref_index":57,"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":29,"is_internal_anchor":false},{"citing_arxiv_id":"2509.18052","citing_title":"The PIMMUR Principles: Ensuring Validity in Collective Behavior of LLM Societies","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2311.12983","citing_title":"GAIA: a benchmark for General AI Assistants","ref_index":141,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ABTUNJ656YVCHXA6KQA37XHBAI","json":"https://pith.science/pith/ABTUNJ656YVCHXA6KQA37XHBAI.json","graph_json":"https://pith.science/api/pith-number/ABTUNJ656YVCHXA6KQA37XHBAI/graph.json","events_json":"https://pith.science/api/pith-number/ABTUNJ656YVCHXA6KQA37XHBAI/events.json","paper":"https://pith.science/paper/ABTUNJ65"},"agent_actions":{"view_html":"https://pith.science/pith/ABTUNJ656YVCHXA6KQA37XHBAI","download_json":"https://pith.science/pith/ABTUNJ656YVCHXA6KQA37XHBAI.json","view_paper":"https://pith.science/paper/ABTUNJ65","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2302.08399&json=true","fetch_graph":"https://pith.science/api/pith-number/ABTUNJ656YVCHXA6KQA37XHBAI/graph.json","fetch_events":"https://pith.science/api/pith-number/ABTUNJ656YVCHXA6KQA37XHBAI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ABTUNJ656YVCHXA6KQA37XHBAI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ABTUNJ656YVCHXA6KQA37XHBAI/action/storage_attestation","attest_author":"https://pith.science/pith/ABTUNJ656YVCHXA6KQA37XHBAI/action/author_attestation","sign_citation":"https://pith.science/pith/ABTUNJ656YVCHXA6KQA37XHBAI/action/citation_signature","submit_replication":"https://pith.science/pith/ABTUNJ656YVCHXA6KQA37XHBAI/action/replication_record"}},"created_at":"2026-07-05T05:51:02.365353+00:00","updated_at":"2026-07-05T05:51:02.365353+00:00"}