{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:TBVBJKTJSKJHUDSOW745GOXBS2","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"dc8b53643d683520b46449e0d015fbdc3cf1b3f701da222ae712c3432a327afe","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-28T14:36:57Z","title_canon_sha256":"cecf73f352071d0d7e1352f4a5b3fb2274375dad7e3750a55f93fa8badf81fb7"},"schema_version":"1.0","source":{"id":"2502.21098","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.21098","created_at":"2026-07-05T10:21:44Z"},{"alias_kind":"arxiv_version","alias_value":"2502.21098v1","created_at":"2026-07-05T10:21:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.21098","created_at":"2026-07-05T10:21:44Z"},{"alias_kind":"pith_short_12","alias_value":"TBVBJKTJSKJH","created_at":"2026-07-05T10:21:44Z"},{"alias_kind":"pith_short_16","alias_value":"TBVBJKTJSKJHUDSO","created_at":"2026-07-05T10:21:44Z"},{"alias_kind":"pith_short_8","alias_value":"TBVBJKTJ","created_at":"2026-07-05T10:21:44Z"}],"graph_snapshots":[{"event_id":"sha256:164230175f2ef66f60e771a822ee7a14b2b6fed6677e0909325e1460a4326500","target":"graph","created_at":"2026-07-05T10:21:44Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2502.21098/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The question of whether large language models (LLMs) possess Theory of Mind (ToM) -- often defined as the ability to reason about others' mental states -- has sparked significant scientific and public interest. However, the evidence as to whether LLMs possess ToM is mixed, and the recent growth in evaluations has not resulted in a convergence. Here, we take inspiration from cognitive science to re-evaluate the state of ToM evaluation in LLMs. We argue that a major reason for the disagreement on whether LLMs have ToM is a lack of clarity on whether models should be expected to match human behav","authors_text":"Felix Sosa, Jennifer Hu, Tomer Ullman","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-28T14:36:57Z","title":"Re-evaluating Theory of Mind evaluation in large language models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.21098","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:ccc469b3947187f5619b03f0afe810397f3dc5c47a201f134d69c3bf57c9bf78","target":"record","created_at":"2026-07-05T10:21:44Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"dc8b53643d683520b46449e0d015fbdc3cf1b3f701da222ae712c3432a327afe","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-28T14:36:57Z","title_canon_sha256":"cecf73f352071d0d7e1352f4a5b3fb2274375dad7e3750a55f93fa8badf81fb7"},"schema_version":"1.0","source":{"id":"2502.21098","kind":"arxiv","version":1}},"canonical_sha256":"986a14aa6992927a0e4eb7f9d33ae196be6eb0a87ddafd59643fd21443dbfee8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"986a14aa6992927a0e4eb7f9d33ae196be6eb0a87ddafd59643fd21443dbfee8","first_computed_at":"2026-07-05T10:21:44.283448Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:21:44.283448Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"oZh3CK898sgEB3lyS7oT6Q2egQoGPf0escZBwnofwc2l6/Le96yh1hp9+IR1RjXN3GNPVOInpczEoPGueQJ5Aw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:21:44.283960Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.21098","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ccc469b3947187f5619b03f0afe810397f3dc5c47a201f134d69c3bf57c9bf78","sha256:164230175f2ef66f60e771a822ee7a14b2b6fed6677e0909325e1460a4326500"],"state_sha256":"c3b8445030263bd8c36325edf3b2d42125f29faf6a9d5f378fcdfc0de5bc1557"}