{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:XLAAXLMKBDUR5N5OCFRATPXMJP","short_pith_number":"pith:XLAAXLMK","schema_version":"1.0","canonical_sha256":"bac00bad8a08e91eb7ae116209beec4bf2522da300c6811b6f5613d65f8c7926","source":{"kind":"arxiv","id":"2504.01698","version":3},"attestation_state":"computed","paper":{"title":"Do Theory of Mind Benchmarks Need Explicit Human-like Reasoning in Language Models?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chunhui Zhang, Jiajun Song, Lifeng Fan, Wei Wang, Yi-Long Lu","submitted_at":"2025-04-02T12:58:42Z","abstract_excerpt":"Theory of Mind (ToM), the ability to attribute mental states to others, is fundamental for human social intelligence and a critical capability for advanced Artificial Intelligence. Recent advancements in Large Language Models (LLMs) have shown promising performance on ToM benchmarks, raising the question: Do these benchmarks necessitate explicit human-like reasoning processes, or can models succeed through alternative strategies? We investigate this question empirically by applying Reinforcement Learning (RL) and Supervised Fine-Tuning (SFT) to LLMs of varying scales (0.5B to 7B parameters) an"},"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":"2504.01698","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-02T12:58:42Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"e35b9cacee8d7cd56ce25e939c00b263eb47d565ab594d4937e45560e0cbbfc9","abstract_canon_sha256":"a8506f61b536b86ad424ca88af62943ab1ed4d729c0d88fcfd82998d2899c0fb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:03:58.937620Z","signature_b64":"1ev3md7QxkZrbvlX7Nl4sk8wtEmEp+ckq46XQ7uVh/39D6BgJ69sWnMXrpE5o+wG/k0EVoDryenJYnNIa0gbAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bac00bad8a08e91eb7ae116209beec4bf2522da300c6811b6f5613d65f8c7926","last_reissued_at":"2026-07-05T11:03:58.937129Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:03:58.937129Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Do Theory of Mind Benchmarks Need Explicit Human-like Reasoning in Language Models?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chunhui Zhang, Jiajun Song, Lifeng Fan, Wei Wang, Yi-Long Lu","submitted_at":"2025-04-02T12:58:42Z","abstract_excerpt":"Theory of Mind (ToM), the ability to attribute mental states to others, is fundamental for human social intelligence and a critical capability for advanced Artificial Intelligence. Recent advancements in Large Language Models (LLMs) have shown promising performance on ToM benchmarks, raising the question: Do these benchmarks necessitate explicit human-like reasoning processes, or can models succeed through alternative strategies? We investigate this question empirically by applying Reinforcement Learning (RL) and Supervised Fine-Tuning (SFT) to LLMs of varying scales (0.5B to 7B parameters) an"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.01698","kind":"arxiv","version":3},"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/2504.01698/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":"2504.01698","created_at":"2026-07-05T11:03:58.937186+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.01698v3","created_at":"2026-07-05T11:03:58.937186+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.01698","created_at":"2026-07-05T11:03:58.937186+00:00"},{"alias_kind":"pith_short_12","alias_value":"XLAAXLMKBDUR","created_at":"2026-07-05T11:03:58.937186+00:00"},{"alias_kind":"pith_short_16","alias_value":"XLAAXLMKBDUR5N5O","created_at":"2026-07-05T11:03:58.937186+00:00"},{"alias_kind":"pith_short_8","alias_value":"XLAAXLMK","created_at":"2026-07-05T11:03:58.937186+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"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":78,"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":96,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00240","citing_title":"MindZero: Learning Online Mental Reasoning With Zero Annotations","ref_index":67,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15205","citing_title":"Does Theory of Mind Improvement Really Benefit Human-AI Interactions? Empirical Findings from Interactive Evaluations","ref_index":27,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XLAAXLMKBDUR5N5OCFRATPXMJP","json":"https://pith.science/pith/XLAAXLMKBDUR5N5OCFRATPXMJP.json","graph_json":"https://pith.science/api/pith-number/XLAAXLMKBDUR5N5OCFRATPXMJP/graph.json","events_json":"https://pith.science/api/pith-number/XLAAXLMKBDUR5N5OCFRATPXMJP/events.json","paper":"https://pith.science/paper/XLAAXLMK"},"agent_actions":{"view_html":"https://pith.science/pith/XLAAXLMKBDUR5N5OCFRATPXMJP","download_json":"https://pith.science/pith/XLAAXLMKBDUR5N5OCFRATPXMJP.json","view_paper":"https://pith.science/paper/XLAAXLMK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.01698&json=true","fetch_graph":"https://pith.science/api/pith-number/XLAAXLMKBDUR5N5OCFRATPXMJP/graph.json","fetch_events":"https://pith.science/api/pith-number/XLAAXLMKBDUR5N5OCFRATPXMJP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XLAAXLMKBDUR5N5OCFRATPXMJP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XLAAXLMKBDUR5N5OCFRATPXMJP/action/storage_attestation","attest_author":"https://pith.science/pith/XLAAXLMKBDUR5N5OCFRATPXMJP/action/author_attestation","sign_citation":"https://pith.science/pith/XLAAXLMKBDUR5N5OCFRATPXMJP/action/citation_signature","submit_replication":"https://pith.science/pith/XLAAXLMKBDUR5N5OCFRATPXMJP/action/replication_record"}},"created_at":"2026-07-05T11:03:58.937186+00:00","updated_at":"2026-07-05T11:03:58.937186+00:00"}