{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:QDYELRVWUJMZYCG6JF43CT4WLR","short_pith_number":"pith:QDYELRVW","canonical_record":{"source":{"id":"2305.13534","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-22T23:14:28Z","cross_cats_sorted":[],"title_canon_sha256":"a3bdcfbff37c143182479f820162f1554b04f9e72ade1e02bf667ef75dacc92d","abstract_canon_sha256":"06043fd402ff979d903403b52f8e1cd6cf9cf41758f3813c7e0d14ca92a2b6a7"},"schema_version":"1.0"},"canonical_sha256":"80f045c6b6a2599c08de4979b14f965c5e4c9447ee4b1d55667266143791c9c8","source":{"kind":"arxiv","id":"2305.13534","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.13534","created_at":"2026-07-05T06:12:53Z"},{"alias_kind":"arxiv_version","alias_value":"2305.13534v1","created_at":"2026-07-05T06:12:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.13534","created_at":"2026-07-05T06:12:53Z"},{"alias_kind":"pith_short_12","alias_value":"QDYELRVWUJMZ","created_at":"2026-07-05T06:12:53Z"},{"alias_kind":"pith_short_16","alias_value":"QDYELRVWUJMZYCG6","created_at":"2026-07-05T06:12:53Z"},{"alias_kind":"pith_short_8","alias_value":"QDYELRVW","created_at":"2026-07-05T06:12:53Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:QDYELRVWUJMZYCG6JF43CT4WLR","target":"record","payload":{"canonical_record":{"source":{"id":"2305.13534","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-22T23:14:28Z","cross_cats_sorted":[],"title_canon_sha256":"a3bdcfbff37c143182479f820162f1554b04f9e72ade1e02bf667ef75dacc92d","abstract_canon_sha256":"06043fd402ff979d903403b52f8e1cd6cf9cf41758f3813c7e0d14ca92a2b6a7"},"schema_version":"1.0"},"canonical_sha256":"80f045c6b6a2599c08de4979b14f965c5e4c9447ee4b1d55667266143791c9c8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:12:53.956114Z","signature_b64":"cbvr9cTVxgMx4s9WKEo+It8ez8KddTKkuL1F0c0IbGCiIpMZKP3FOThGiDGKRIrnwMtIneh1rQNpbvq9PruuBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"80f045c6b6a2599c08de4979b14f965c5e4c9447ee4b1d55667266143791c9c8","last_reissued_at":"2026-07-05T06:12:53.955669Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:12:53.955669Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2305.13534","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":"2026-07-05T06:12:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Y+2anFWup7vitD6NKITP3rbI3KyQugoiqnHHuxoA7yPrP07VmyItH/Y78szhbIA1iZRWNB/Rm6Cdfa3DawIOCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T19:51:28.493828Z"},"content_sha256":"4bacbf13ac53a87683cb60c36805e15e6f82a45dabeedd54efe7f64dbad7d7c8","schema_version":"1.0","event_id":"sha256:4bacbf13ac53a87683cb60c36805e15e6f82a45dabeedd54efe7f64dbad7d7c8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:QDYELRVWUJMZYCG6JF43CT4WLR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"How Language Model Hallucinations Can Snowball","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alisa Liu, Muru Zhang, Noah A. Smith, Ofir Press, William Merrill","submitted_at":"2023-05-22T23:14:28Z","abstract_excerpt":"A major risk of using language models in practical applications is their tendency to hallucinate incorrect statements. Hallucinations are often attributed to knowledge gaps in LMs, but we hypothesize that in some cases, when justifying previously generated hallucinations, LMs output false claims that they can separately recognize as incorrect. We construct three question-answering datasets where ChatGPT and GPT-4 often state an incorrect answer and offer an explanation with at least one incorrect claim. Crucially, we find that ChatGPT and GPT-4 can identify 67% and 87% of their own mistakes, r"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.13534","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/2305.13534/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":"2026-07-05T06:12:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FAle3OlSmilvLK44ZRD4V9xVcYHFwfYuw5EFdbl/lDBai+FJQEbZIb+yI4sXLKEIV59p2YaHJH/knevD3WHIAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T19:51:28.494455Z"},"content_sha256":"a2700d368c23855e1c94e2b9a46b62873e1b2e63a8eaf00b2e9c5bc249eee413","schema_version":"1.0","event_id":"sha256:a2700d368c23855e1c94e2b9a46b62873e1b2e63a8eaf00b2e9c5bc249eee413"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QDYELRVWUJMZYCG6JF43CT4WLR/bundle.json","state_url":"https://pith.science/pith/QDYELRVWUJMZYCG6JF43CT4WLR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QDYELRVWUJMZYCG6JF43CT4WLR/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":"2026-08-03T19:51:28Z","links":{"resolver":"https://pith.science/pith/QDYELRVWUJMZYCG6JF43CT4WLR","bundle":"https://pith.science/pith/QDYELRVWUJMZYCG6JF43CT4WLR/bundle.json","state":"https://pith.science/pith/QDYELRVWUJMZYCG6JF43CT4WLR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QDYELRVWUJMZYCG6JF43CT4WLR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:QDYELRVWUJMZYCG6JF43CT4WLR","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":"06043fd402ff979d903403b52f8e1cd6cf9cf41758f3813c7e0d14ca92a2b6a7","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-22T23:14:28Z","title_canon_sha256":"a3bdcfbff37c143182479f820162f1554b04f9e72ade1e02bf667ef75dacc92d"},"schema_version":"1.0","source":{"id":"2305.13534","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.13534","created_at":"2026-07-05T06:12:53Z"},{"alias_kind":"arxiv_version","alias_value":"2305.13534v1","created_at":"2026-07-05T06:12:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.13534","created_at":"2026-07-05T06:12:53Z"},{"alias_kind":"pith_short_12","alias_value":"QDYELRVWUJMZ","created_at":"2026-07-05T06:12:53Z"},{"alias_kind":"pith_short_16","alias_value":"QDYELRVWUJMZYCG6","created_at":"2026-07-05T06:12:53Z"},{"alias_kind":"pith_short_8","alias_value":"QDYELRVW","created_at":"2026-07-05T06:12:53Z"}],"graph_snapshots":[{"event_id":"sha256:a2700d368c23855e1c94e2b9a46b62873e1b2e63a8eaf00b2e9c5bc249eee413","target":"graph","created_at":"2026-07-05T06:12:53Z","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/2305.13534/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"A major risk of using language models in practical applications is their tendency to hallucinate incorrect statements. Hallucinations are often attributed to knowledge gaps in LMs, but we hypothesize that in some cases, when justifying previously generated hallucinations, LMs output false claims that they can separately recognize as incorrect. We construct three question-answering datasets where ChatGPT and GPT-4 often state an incorrect answer and offer an explanation with at least one incorrect claim. Crucially, we find that ChatGPT and GPT-4 can identify 67% and 87% of their own mistakes, r","authors_text":"Alisa Liu, Muru Zhang, Noah A. Smith, Ofir Press, William Merrill","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-22T23:14:28Z","title":"How Language Model Hallucinations Can Snowball"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.13534","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":"sha256:4bacbf13ac53a87683cb60c36805e15e6f82a45dabeedd54efe7f64dbad7d7c8","target":"record","created_at":"2026-07-05T06:12:53Z","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":"06043fd402ff979d903403b52f8e1cd6cf9cf41758f3813c7e0d14ca92a2b6a7","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-22T23:14:28Z","title_canon_sha256":"a3bdcfbff37c143182479f820162f1554b04f9e72ade1e02bf667ef75dacc92d"},"schema_version":"1.0","source":{"id":"2305.13534","kind":"arxiv","version":1}},"canonical_sha256":"80f045c6b6a2599c08de4979b14f965c5e4c9447ee4b1d55667266143791c9c8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"80f045c6b6a2599c08de4979b14f965c5e4c9447ee4b1d55667266143791c9c8","first_computed_at":"2026-07-05T06:12:53.955669Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:12:53.955669Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"cbvr9cTVxgMx4s9WKEo+It8ez8KddTKkuL1F0c0IbGCiIpMZKP3FOThGiDGKRIrnwMtIneh1rQNpbvq9PruuBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T06:12:53.956114Z","signed_message":"canonical_sha256_bytes"},"source_id":"2305.13534","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4bacbf13ac53a87683cb60c36805e15e6f82a45dabeedd54efe7f64dbad7d7c8","sha256:a2700d368c23855e1c94e2b9a46b62873e1b2e63a8eaf00b2e9c5bc249eee413"],"state_sha256":"65322a00244df177609b8625360cc4526851219eae813efec74788e727c18a81"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cjpImWTKpgswpIpyGhwhwOesC37gObRNxwajC+HoSZpnoz3Pu5mlnw4MYjP1mcmbgLsFMO+qqVmQCAypW0hVAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T19:51:28.498878Z","bundle_sha256":"a7874ac2c9d5d32a67e80310d73c91e42726246499a85757d1b0334adf043039"}}