{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:XX3YFJ72DKPJKWAIBRPFXWQWDF","short_pith_number":"pith:XX3YFJ72","canonical_record":{"source":{"id":"2305.14763","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-24T06:14:31Z","cross_cats_sorted":[],"title_canon_sha256":"9c93b4b93b20a8f2d69997227ef33c658bb8a1d22cc86db073d200af6242cd28","abstract_canon_sha256":"89347a6dad11c48cd47739b0a911ccf69fa633c80890365b0b96ca3ebf35e34d"},"schema_version":"1.0"},"canonical_sha256":"bdf782a7fa1a9e9558080c5e5bda16196b965de3438c6218994a36e436a9c85c","source":{"kind":"arxiv","id":"2305.14763","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.14763","created_at":"2026-07-05T06:13:27Z"},{"alias_kind":"arxiv_version","alias_value":"2305.14763v1","created_at":"2026-07-05T06:13:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.14763","created_at":"2026-07-05T06:13:27Z"},{"alias_kind":"pith_short_12","alias_value":"XX3YFJ72DKPJ","created_at":"2026-07-05T06:13:27Z"},{"alias_kind":"pith_short_16","alias_value":"XX3YFJ72DKPJKWAI","created_at":"2026-07-05T06:13:27Z"},{"alias_kind":"pith_short_8","alias_value":"XX3YFJ72","created_at":"2026-07-05T06:13:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:XX3YFJ72DKPJKWAIBRPFXWQWDF","target":"record","payload":{"canonical_record":{"source":{"id":"2305.14763","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-24T06:14:31Z","cross_cats_sorted":[],"title_canon_sha256":"9c93b4b93b20a8f2d69997227ef33c658bb8a1d22cc86db073d200af6242cd28","abstract_canon_sha256":"89347a6dad11c48cd47739b0a911ccf69fa633c80890365b0b96ca3ebf35e34d"},"schema_version":"1.0"},"canonical_sha256":"bdf782a7fa1a9e9558080c5e5bda16196b965de3438c6218994a36e436a9c85c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:13:27.692695Z","signature_b64":"xu/00WZjpvA8kVv1Y9QZfnXkyF+NBX8f+GY+Act2cbXqHCuuU1ZDO5nA6ExIoMQrTFQFW7KCumDiaFRnLkdYCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bdf782a7fa1a9e9558080c5e5bda16196b965de3438c6218994a36e436a9c85c","last_reissued_at":"2026-07-05T06:13:27.692250Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:13:27.692250Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2305.14763","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T06:13:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BvFpEKoKqXwKLWqnVIeUDxbcrdirUFo7t/koHA4YJ0Jwx89uEp73f8Bc2WgaxwyYuAovXXbINpaQY4yr5of1Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-26T01:46:48.389023Z"},"content_sha256":"fc78a82e55cb34a4d552b2eba4722815d874778fc1c18195191e37b54ad0f0d5","schema_version":"1.0","event_id":"sha256:fc78a82e55cb34a4d552b2eba4722815d874778fc1c18195191e37b54ad0f0d5"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:XX3YFJ72DKPJKWAIBRPFXWQWDF","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Clever Hans or Neural Theory of Mind? Stress Testing Social Reasoning in Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Maarten Sap, Mosh Levy, Natalie Shapira, Seyed Hossein Alavi, Vered Shwartz, Xuhui Zhou, Yejin Choi, Yoav Goldberg","submitted_at":"2023-05-24T06:14:31Z","abstract_excerpt":"The escalating debate on AI's capabilities warrants developing reliable metrics to assess machine \"intelligence\". Recently, many anecdotal examples were used to suggest that newer large language models (LLMs) like ChatGPT and GPT-4 exhibit Neural Theory-of-Mind (N-ToM); however, prior work reached conflicting conclusions regarding those abilities. We investigate the extent of LLMs' N-ToM through an extensive evaluation on 6 tasks and find that while LLMs exhibit certain N-ToM abilities, this behavior is far from being robust. We further examine the factors impacting performance on N-ToM tasks "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.14763","kind":"arxiv","version":1},"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/2305.14763/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":"2026-07-05T06:13:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ykZOv49zeStJ2Tu+iS2hj0Fnax5r9hLiEcgjXYMPm5xJbcRr6UqbAJS6VUzsOM6RIFGI0gcLRSSQUAhg0GB3CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-26T01:46:48.389400Z"},"content_sha256":"41efee2e25cb418c41d75bf7d9e8142af4a9df5401fe95a2cd06806587086823","schema_version":"1.0","event_id":"sha256:41efee2e25cb418c41d75bf7d9e8142af4a9df5401fe95a2cd06806587086823"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XX3YFJ72DKPJKWAIBRPFXWQWDF/bundle.json","state_url":"https://pith.science/pith/XX3YFJ72DKPJKWAIBRPFXWQWDF/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XX3YFJ72DKPJKWAIBRPFXWQWDF/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":"2026-07-26T01:46:48Z","links":{"resolver":"https://pith.science/pith/XX3YFJ72DKPJKWAIBRPFXWQWDF","bundle":"https://pith.science/pith/XX3YFJ72DKPJKWAIBRPFXWQWDF/bundle.json","state":"https://pith.science/pith/XX3YFJ72DKPJKWAIBRPFXWQWDF/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XX3YFJ72DKPJKWAIBRPFXWQWDF/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:XX3YFJ72DKPJKWAIBRPFXWQWDF","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":"89347a6dad11c48cd47739b0a911ccf69fa633c80890365b0b96ca3ebf35e34d","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-24T06:14:31Z","title_canon_sha256":"9c93b4b93b20a8f2d69997227ef33c658bb8a1d22cc86db073d200af6242cd28"},"schema_version":"1.0","source":{"id":"2305.14763","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.14763","created_at":"2026-07-05T06:13:27Z"},{"alias_kind":"arxiv_version","alias_value":"2305.14763v1","created_at":"2026-07-05T06:13:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.14763","created_at":"2026-07-05T06:13:27Z"},{"alias_kind":"pith_short_12","alias_value":"XX3YFJ72DKPJ","created_at":"2026-07-05T06:13:27Z"},{"alias_kind":"pith_short_16","alias_value":"XX3YFJ72DKPJKWAI","created_at":"2026-07-05T06:13:27Z"},{"alias_kind":"pith_short_8","alias_value":"XX3YFJ72","created_at":"2026-07-05T06:13:27Z"}],"graph_snapshots":[{"event_id":"sha256:41efee2e25cb418c41d75bf7d9e8142af4a9df5401fe95a2cd06806587086823","target":"graph","created_at":"2026-07-05T06:13:27Z","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/2305.14763/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The escalating debate on AI's capabilities warrants developing reliable metrics to assess machine \"intelligence\". Recently, many anecdotal examples were used to suggest that newer large language models (LLMs) like ChatGPT and GPT-4 exhibit Neural Theory-of-Mind (N-ToM); however, prior work reached conflicting conclusions regarding those abilities. We investigate the extent of LLMs' N-ToM through an extensive evaluation on 6 tasks and find that while LLMs exhibit certain N-ToM abilities, this behavior is far from being robust. We further examine the factors impacting performance on N-ToM tasks ","authors_text":"Maarten Sap, Mosh Levy, Natalie Shapira, Seyed Hossein Alavi, Vered Shwartz, Xuhui Zhou, Yejin Choi, Yoav Goldberg","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-24T06:14:31Z","title":"Clever Hans or Neural Theory of Mind? Stress Testing Social Reasoning in Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.14763","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:fc78a82e55cb34a4d552b2eba4722815d874778fc1c18195191e37b54ad0f0d5","target":"record","created_at":"2026-07-05T06:13:27Z","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":"89347a6dad11c48cd47739b0a911ccf69fa633c80890365b0b96ca3ebf35e34d","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-24T06:14:31Z","title_canon_sha256":"9c93b4b93b20a8f2d69997227ef33c658bb8a1d22cc86db073d200af6242cd28"},"schema_version":"1.0","source":{"id":"2305.14763","kind":"arxiv","version":1}},"canonical_sha256":"bdf782a7fa1a9e9558080c5e5bda16196b965de3438c6218994a36e436a9c85c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bdf782a7fa1a9e9558080c5e5bda16196b965de3438c6218994a36e436a9c85c","first_computed_at":"2026-07-05T06:13:27.692250Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:13:27.692250Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"xu/00WZjpvA8kVv1Y9QZfnXkyF+NBX8f+GY+Act2cbXqHCuuU1ZDO5nA6ExIoMQrTFQFW7KCumDiaFRnLkdYCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T06:13:27.692695Z","signed_message":"canonical_sha256_bytes"},"source_id":"2305.14763","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:fc78a82e55cb34a4d552b2eba4722815d874778fc1c18195191e37b54ad0f0d5","sha256:41efee2e25cb418c41d75bf7d9e8142af4a9df5401fe95a2cd06806587086823"],"state_sha256":"4251a54944a4c292213a88977b84507df6cded1eeb30a92d683c0221ab1cf88c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Zx6sdAlpCI8DxBIrchqECqFa7U/05ngL8PhHzoNmEPPyWsCHauxd+lJaCQ/SmsOwarXvotHVVV+CfXdtE7HqBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-26T01:46:48.393539Z","bundle_sha256":"dca66fc18ae9aa89c29f045a4515d86f1c1d253a29d7a1a1bc554dbc24aed47c"}}