{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:CFGRQ7U55QVZ2JR4HOWM37KTPW","short_pith_number":"pith:CFGRQ7U5","canonical_record":{"source":{"id":"2402.06044","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-02-08T20:35:06Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"23cd939941b65b5ef926cab9b8b7159e870f6a0390d205cba1a800ac7195dbc1","abstract_canon_sha256":"687e051a7284996a7367f639decc3ba94ae87c89093b6093c9768a2a4c38d00c"},"schema_version":"1.0"},"canonical_sha256":"114d187e9dec2b9d263c3baccdfd537dbf0cb0063e722777fe9b7af451b36e9b","source":{"kind":"arxiv","id":"2402.06044","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.06044","created_at":"2026-07-05T08:26:19Z"},{"alias_kind":"arxiv_version","alias_value":"2402.06044v3","created_at":"2026-07-05T08:26:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.06044","created_at":"2026-07-05T08:26:19Z"},{"alias_kind":"pith_short_12","alias_value":"CFGRQ7U55QVZ","created_at":"2026-07-05T08:26:19Z"},{"alias_kind":"pith_short_16","alias_value":"CFGRQ7U55QVZ2JR4","created_at":"2026-07-05T08:26:19Z"},{"alias_kind":"pith_short_8","alias_value":"CFGRQ7U5","created_at":"2026-07-05T08:26:19Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:CFGRQ7U55QVZ2JR4HOWM37KTPW","target":"record","payload":{"canonical_record":{"source":{"id":"2402.06044","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-02-08T20:35:06Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"23cd939941b65b5ef926cab9b8b7159e870f6a0390d205cba1a800ac7195dbc1","abstract_canon_sha256":"687e051a7284996a7367f639decc3ba94ae87c89093b6093c9768a2a4c38d00c"},"schema_version":"1.0"},"canonical_sha256":"114d187e9dec2b9d263c3baccdfd537dbf0cb0063e722777fe9b7af451b36e9b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:26:19.160251Z","signature_b64":"V3REaTH2C5Ev5Al5y/onQrG+5LwYJCyJfkOE/yTtDcOx7AWJD8kD1fIZDe9w1OkPy5MOnDFWTAKxxhottxzOBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"114d187e9dec2b9d263c3baccdfd537dbf0cb0063e722777fe9b7af451b36e9b","last_reissued_at":"2026-07-05T08:26:19.159753Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:26:19.159753Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.06044","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:26:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FyTyL41CRTgHvMuSlbi0kDSBoO4igRGvtx94fm3p18rdeNTa9bleTTnSAnqXON57rArWteM3/fNcKWqnYEr9BA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T22:30:48.898725Z"},"content_sha256":"9a549d7ae1adb4df7b8f602c93c595819586352d107d6297162ea4602648006d","schema_version":"1.0","event_id":"sha256:9a549d7ae1adb4df7b8f602c93c595819586352d107d6297162ea4602648006d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:CFGRQ7U55QVZ2JR4HOWM37KTPW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"OpenToM: A Comprehensive Benchmark for Evaluating Theory-of-Mind Reasoning Capabilities of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Hainiu Xu, Jinhua Du, Lixing Zhu, Runcong Zhao, Yulan He","submitted_at":"2024-02-08T20:35:06Z","abstract_excerpt":"Neural Theory-of-Mind (N-ToM), machine's ability to understand and keep track of the mental states of others, is pivotal in developing socially intelligent agents. However, prevalent N-ToM benchmarks have several shortcomings, including the presence of ambiguous and artificial narratives, absence of personality traits and preferences, a lack of questions addressing characters' psychological mental states, and limited diversity in the questions posed. In response to these issues, we construct OpenToM, a new benchmark for assessing N-ToM with (1) longer and clearer narrative stories, (2) charact"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.06044","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/2402.06044/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:26:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"11gZFh0VQr8FQbEmKJqmedQjwxRBtqQ5VV/JlpOK/dmtXgbXbVQ1y1GrQ+95YRr+kI8EPVdC71kkE0P0N5tWDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T22:30:48.899233Z"},"content_sha256":"c6808968afee6b4d72bcd79792e6d65aac59a09f13606922791af70488cbda47","schema_version":"1.0","event_id":"sha256:c6808968afee6b4d72bcd79792e6d65aac59a09f13606922791af70488cbda47"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CFGRQ7U55QVZ2JR4HOWM37KTPW/bundle.json","state_url":"https://pith.science/pith/CFGRQ7U55QVZ2JR4HOWM37KTPW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CFGRQ7U55QVZ2JR4HOWM37KTPW/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-07T22:30:48Z","links":{"resolver":"https://pith.science/pith/CFGRQ7U55QVZ2JR4HOWM37KTPW","bundle":"https://pith.science/pith/CFGRQ7U55QVZ2JR4HOWM37KTPW/bundle.json","state":"https://pith.science/pith/CFGRQ7U55QVZ2JR4HOWM37KTPW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CFGRQ7U55QVZ2JR4HOWM37KTPW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:CFGRQ7U55QVZ2JR4HOWM37KTPW","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":"687e051a7284996a7367f639decc3ba94ae87c89093b6093c9768a2a4c38d00c","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-02-08T20:35:06Z","title_canon_sha256":"23cd939941b65b5ef926cab9b8b7159e870f6a0390d205cba1a800ac7195dbc1"},"schema_version":"1.0","source":{"id":"2402.06044","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.06044","created_at":"2026-07-05T08:26:19Z"},{"alias_kind":"arxiv_version","alias_value":"2402.06044v3","created_at":"2026-07-05T08:26:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.06044","created_at":"2026-07-05T08:26:19Z"},{"alias_kind":"pith_short_12","alias_value":"CFGRQ7U55QVZ","created_at":"2026-07-05T08:26:19Z"},{"alias_kind":"pith_short_16","alias_value":"CFGRQ7U55QVZ2JR4","created_at":"2026-07-05T08:26:19Z"},{"alias_kind":"pith_short_8","alias_value":"CFGRQ7U5","created_at":"2026-07-05T08:26:19Z"}],"graph_snapshots":[{"event_id":"sha256:c6808968afee6b4d72bcd79792e6d65aac59a09f13606922791af70488cbda47","target":"graph","created_at":"2026-07-05T08:26:19Z","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/2402.06044/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Neural Theory-of-Mind (N-ToM), machine's ability to understand and keep track of the mental states of others, is pivotal in developing socially intelligent agents. However, prevalent N-ToM benchmarks have several shortcomings, including the presence of ambiguous and artificial narratives, absence of personality traits and preferences, a lack of questions addressing characters' psychological mental states, and limited diversity in the questions posed. In response to these issues, we construct OpenToM, a new benchmark for assessing N-ToM with (1) longer and clearer narrative stories, (2) charact","authors_text":"Hainiu Xu, Jinhua Du, Lixing Zhu, Runcong Zhao, Yulan He","cross_cats":["cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-02-08T20:35:06Z","title":"OpenToM: A Comprehensive Benchmark for Evaluating Theory-of-Mind Reasoning Capabilities of Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.06044","kind":"arxiv","version":3},"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":"sha256:9a549d7ae1adb4df7b8f602c93c595819586352d107d6297162ea4602648006d","target":"record","created_at":"2026-07-05T08:26:19Z","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":"687e051a7284996a7367f639decc3ba94ae87c89093b6093c9768a2a4c38d00c","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-02-08T20:35:06Z","title_canon_sha256":"23cd939941b65b5ef926cab9b8b7159e870f6a0390d205cba1a800ac7195dbc1"},"schema_version":"1.0","source":{"id":"2402.06044","kind":"arxiv","version":3}},"canonical_sha256":"114d187e9dec2b9d263c3baccdfd537dbf0cb0063e722777fe9b7af451b36e9b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"114d187e9dec2b9d263c3baccdfd537dbf0cb0063e722777fe9b7af451b36e9b","first_computed_at":"2026-07-05T08:26:19.159753Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:26:19.159753Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"V3REaTH2C5Ev5Al5y/onQrG+5LwYJCyJfkOE/yTtDcOx7AWJD8kD1fIZDe9w1OkPy5MOnDFWTAKxxhottxzOBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:26:19.160251Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.06044","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9a549d7ae1adb4df7b8f602c93c595819586352d107d6297162ea4602648006d","sha256:c6808968afee6b4d72bcd79792e6d65aac59a09f13606922791af70488cbda47"],"state_sha256":"7148edde424c15e7c6a0f32e7cb3f9aed5dbb631c71bba1ccf3e321a754c551b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SEWv2g/mLjo1+ztHATXtAbc9MC/x3quNSjDGme+cvmQRehx6ITfT8EHMblf6EaK8ZgVJ/FwyR4Q6rwjTDhDPAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T22:30:48.904790Z","bundle_sha256":"aef707175e1479d688b6d315ac82c2bbde9833537323d616a658f9684479c9a0"}}