{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:6DU5NEZ7RDIG5QO5TW5K7SQ6SY","short_pith_number":"pith:6DU5NEZ7","canonical_record":{"source":{"id":"2402.10412","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-16T02:32:06Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"b66e3747c3966904f94f7c2794b088b30a85f7614202923f8449dadfb63fa8c0","abstract_canon_sha256":"2dad6bb735712728a40ecc6a7ced6f2434895aa9c90e5824c84202b4f7bc19fe"},"schema_version":"1.0"},"canonical_sha256":"f0e9d6933f88d06ec1dd9dbaafca1e961d596764f3b1321886bf4b3de711a819","source":{"kind":"arxiv","id":"2402.10412","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.10412","created_at":"2026-07-05T08:28:38Z"},{"alias_kind":"arxiv_version","alias_value":"2402.10412v2","created_at":"2026-07-05T08:28:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.10412","created_at":"2026-07-05T08:28:38Z"},{"alias_kind":"pith_short_12","alias_value":"6DU5NEZ7RDIG","created_at":"2026-07-05T08:28:38Z"},{"alias_kind":"pith_short_16","alias_value":"6DU5NEZ7RDIG5QO5","created_at":"2026-07-05T08:28:38Z"},{"alias_kind":"pith_short_8","alias_value":"6DU5NEZ7","created_at":"2026-07-05T08:28:38Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:6DU5NEZ7RDIG5QO5TW5K7SQ6SY","target":"record","payload":{"canonical_record":{"source":{"id":"2402.10412","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-16T02:32:06Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"b66e3747c3966904f94f7c2794b088b30a85f7614202923f8449dadfb63fa8c0","abstract_canon_sha256":"2dad6bb735712728a40ecc6a7ced6f2434895aa9c90e5824c84202b4f7bc19fe"},"schema_version":"1.0"},"canonical_sha256":"f0e9d6933f88d06ec1dd9dbaafca1e961d596764f3b1321886bf4b3de711a819","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:28:38.197401Z","signature_b64":"DIp8k9LD27O9mj0nxB5bjVI/Ai8psIXtkYtKTVcW5u/ysJjad60tPJVh+93jFJcyPVJ0qmcGrGc1i5vidiezAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f0e9d6933f88d06ec1dd9dbaafca1e961d596764f3b1321886bf4b3de711a819","last_reissued_at":"2026-07-05T08:28:38.196917Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:28:38.196917Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.10412","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:28:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"M92foQq2tcISRlOGtmIkYO28Z85RQDkBnBRqSBfMJADTXl21TOJp6afpbRn1eFv8GsZQSLnuXYGEHRBwNZvFBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T08:34:42.439600Z"},"content_sha256":"b3a682841a64ac853e64de2b33f43ba61cfe3c5c24cd84dc500c6d5323d3891e","schema_version":"1.0","event_id":"sha256:b3a682841a64ac853e64de2b33f43ba61cfe3c5c24cd84dc500c6d5323d3891e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:6DU5NEZ7RDIG5QO5TW5K7SQ6SY","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Measuring and Reducing LLM Hallucination without Gold-Standard Answers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Andrew Estornell, Hongyi Guo, Jean-Francois Ton, Jiaheng Wei, Yang Liu, Yuanshun Yao","submitted_at":"2024-02-16T02:32:06Z","abstract_excerpt":"LLM hallucination, i.e. generating factually incorrect yet seemingly convincing answers, is currently a major threat to the trustworthiness and reliability of LLMs. The first step towards solving this complicated problem is to measure it. However, existing hallucination metrics require having a benchmark dataset with gold-standard answers, i.e. \"best\" or \"correct\" answers written by humans. Such requirements make hallucination measurement costly and prone to human errors. In this work, we propose Factualness Evaluations via Weighting LLMs (FEWL), an innovative hallucination metric that is spec"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.10412","kind":"arxiv","version":2},"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/2402.10412/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":"2026-07-05T08:28:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gM/9Cgu2WmaGET0mB1WHdngMqrkN7Z2u3Km/vK4eNLm+W2dc/YvxxN90Vw7kKxUGpjU/Wm4NWCBi3xGHuuFLBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T08:34:42.440097Z"},"content_sha256":"cbb113cf4a9a23c88cd22c4413ce0f986c60555f2ace961041b45a76d21f7acb","schema_version":"1.0","event_id":"sha256:cbb113cf4a9a23c88cd22c4413ce0f986c60555f2ace961041b45a76d21f7acb"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/6DU5NEZ7RDIG5QO5TW5K7SQ6SY/bundle.json","state_url":"https://pith.science/pith/6DU5NEZ7RDIG5QO5TW5K7SQ6SY/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/6DU5NEZ7RDIG5QO5TW5K7SQ6SY/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":"2026-08-21T08:34:42Z","links":{"resolver":"https://pith.science/pith/6DU5NEZ7RDIG5QO5TW5K7SQ6SY","bundle":"https://pith.science/pith/6DU5NEZ7RDIG5QO5TW5K7SQ6SY/bundle.json","state":"https://pith.science/pith/6DU5NEZ7RDIG5QO5TW5K7SQ6SY/state.json","well_known_bundle":"https://pith.science/.well-known/pith/6DU5NEZ7RDIG5QO5TW5K7SQ6SY/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:6DU5NEZ7RDIG5QO5TW5K7SQ6SY","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":"2dad6bb735712728a40ecc6a7ced6f2434895aa9c90e5824c84202b4f7bc19fe","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-16T02:32:06Z","title_canon_sha256":"b66e3747c3966904f94f7c2794b088b30a85f7614202923f8449dadfb63fa8c0"},"schema_version":"1.0","source":{"id":"2402.10412","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.10412","created_at":"2026-07-05T08:28:38Z"},{"alias_kind":"arxiv_version","alias_value":"2402.10412v2","created_at":"2026-07-05T08:28:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.10412","created_at":"2026-07-05T08:28:38Z"},{"alias_kind":"pith_short_12","alias_value":"6DU5NEZ7RDIG","created_at":"2026-07-05T08:28:38Z"},{"alias_kind":"pith_short_16","alias_value":"6DU5NEZ7RDIG5QO5","created_at":"2026-07-05T08:28:38Z"},{"alias_kind":"pith_short_8","alias_value":"6DU5NEZ7","created_at":"2026-07-05T08:28:38Z"}],"graph_snapshots":[{"event_id":"sha256:cbb113cf4a9a23c88cd22c4413ce0f986c60555f2ace961041b45a76d21f7acb","target":"graph","created_at":"2026-07-05T08:28:38Z","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/2402.10412/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"LLM hallucination, i.e. generating factually incorrect yet seemingly convincing answers, is currently a major threat to the trustworthiness and reliability of LLMs. The first step towards solving this complicated problem is to measure it. However, existing hallucination metrics require having a benchmark dataset with gold-standard answers, i.e. \"best\" or \"correct\" answers written by humans. Such requirements make hallucination measurement costly and prone to human errors. In this work, we propose Factualness Evaluations via Weighting LLMs (FEWL), an innovative hallucination metric that is spec","authors_text":"Andrew Estornell, Hongyi Guo, Jean-Francois Ton, Jiaheng Wei, Yang Liu, Yuanshun Yao","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-16T02:32:06Z","title":"Measuring and Reducing LLM Hallucination without Gold-Standard Answers"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.10412","kind":"arxiv","version":2},"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:b3a682841a64ac853e64de2b33f43ba61cfe3c5c24cd84dc500c6d5323d3891e","target":"record","created_at":"2026-07-05T08:28:38Z","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":"2dad6bb735712728a40ecc6a7ced6f2434895aa9c90e5824c84202b4f7bc19fe","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-16T02:32:06Z","title_canon_sha256":"b66e3747c3966904f94f7c2794b088b30a85f7614202923f8449dadfb63fa8c0"},"schema_version":"1.0","source":{"id":"2402.10412","kind":"arxiv","version":2}},"canonical_sha256":"f0e9d6933f88d06ec1dd9dbaafca1e961d596764f3b1321886bf4b3de711a819","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f0e9d6933f88d06ec1dd9dbaafca1e961d596764f3b1321886bf4b3de711a819","first_computed_at":"2026-07-05T08:28:38.196917Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:28:38.196917Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"DIp8k9LD27O9mj0nxB5bjVI/Ai8psIXtkYtKTVcW5u/ysJjad60tPJVh+93jFJcyPVJ0qmcGrGc1i5vidiezAg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:28:38.197401Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.10412","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b3a682841a64ac853e64de2b33f43ba61cfe3c5c24cd84dc500c6d5323d3891e","sha256:cbb113cf4a9a23c88cd22c4413ce0f986c60555f2ace961041b45a76d21f7acb"],"state_sha256":"f6b42a78a3c2a18bde3c7403d05680366804fa0fa55522110eef2474f1aa6f64"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6ABbQvsdbmvmFv941LL4Ltiz7c7dGKxx9wg95ckTwd/FPTBuhV6vbim4QB0d8VhL3+rQbRy7p01PIdLwcvsFCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-21T08:34:42.443857Z","bundle_sha256":"1ff21dba766723c0ca0d50f7552794543571b843576fd2da18e36949425fc54e"}}