{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:BKGPZG6XQZLZE25PHPIGXEOLZU","short_pith_number":"pith:BKGPZG6X","canonical_record":{"source":{"id":"2310.00741","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-01T17:37:31Z","cross_cats_sorted":[],"title_canon_sha256":"3d7678736fdc7023e8ff56796aef0736c78e2c4e84845c8cb481abbc9a16ef91","abstract_canon_sha256":"567daf31865dcb0ac0d685472a9687b0f0bf65f69397be48d22c0e5e9d60731c"},"schema_version":"1.0"},"canonical_sha256":"0a8cfc9bd78657926baf3bd06b91cbcd317b8275c05d2d8ce08746a26d06493e","source":{"kind":"arxiv","id":"2310.00741","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.00741","created_at":"2026-07-05T07:17:32Z"},{"alias_kind":"arxiv_version","alias_value":"2310.00741v2","created_at":"2026-07-05T07:17:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.00741","created_at":"2026-07-05T07:17:32Z"},{"alias_kind":"pith_short_12","alias_value":"BKGPZG6XQZLZ","created_at":"2026-07-05T07:17:32Z"},{"alias_kind":"pith_short_16","alias_value":"BKGPZG6XQZLZE25P","created_at":"2026-07-05T07:17:32Z"},{"alias_kind":"pith_short_8","alias_value":"BKGPZG6X","created_at":"2026-07-05T07:17:32Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:BKGPZG6XQZLZE25PHPIGXEOLZU","target":"record","payload":{"canonical_record":{"source":{"id":"2310.00741","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-01T17:37:31Z","cross_cats_sorted":[],"title_canon_sha256":"3d7678736fdc7023e8ff56796aef0736c78e2c4e84845c8cb481abbc9a16ef91","abstract_canon_sha256":"567daf31865dcb0ac0d685472a9687b0f0bf65f69397be48d22c0e5e9d60731c"},"schema_version":"1.0"},"canonical_sha256":"0a8cfc9bd78657926baf3bd06b91cbcd317b8275c05d2d8ce08746a26d06493e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:17:32.002206Z","signature_b64":"uNq+wbWmOsDAz6JoKB7J0AhjC1Q/qgJK/JOuyUZzshRqaSxyQpaMRJ+lzbMN0nDUzgz4yXhlUlYnYj8SCrZ0Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0a8cfc9bd78657926baf3bd06b91cbcd317b8275c05d2d8ce08746a26d06493e","last_reissued_at":"2026-07-05T07:17:32.001749Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:17:32.001749Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2310.00741","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":"2026-07-05T07:17:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"QEB1yPFmSSe8eQ3+CL4d/phhrFxG8o/Eo//qdS5dQTeXdszGDTPEEzsxK0nukTlh5XLIldfwiFcPQ7e1bMWvDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T01:22:51.822563Z"},"content_sha256":"fe7907b6d9e6348506987f3c0aea3336ed79a055376e875f96bd7a305bb130c4","schema_version":"1.0","event_id":"sha256:fe7907b6d9e6348506987f3c0aea3336ed79a055376e875f96bd7a305bb130c4"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:BKGPZG6XQZLZE25PHPIGXEOLZU","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"FELM: Benchmarking Factuality Evaluation of Large Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"I-Chun Chern, Jinghan Zhang, Junxian He, Pengfei Liu, Shiqi Chen, Siyang Gao, Yiran Zhao","submitted_at":"2023-10-01T17:37:31Z","abstract_excerpt":"Assessing factuality of text generated by large language models (LLMs) is an emerging yet crucial research area, aimed at alerting users to potential errors and guiding the development of more reliable LLMs. Nonetheless, the evaluators assessing factuality necessitate suitable evaluation themselves to gauge progress and foster advancements. This direction remains under-explored, resulting in substantial impediments to the progress of factuality evaluators. To mitigate this issue, we introduce a benchmark for Factuality Evaluation of large Language Models, referred to as felm. In this benchmark"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.00741","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/2310.00741/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":"2026-07-05T07:17:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6pFLpqmzs6wMhpgExm0wOEsOJCvuDeltbnAVfmyPeF/+l/bLvBpA4jLwHpnE9hI8Y/iQ5CAY5CqDPc+T/+FtDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T01:22:51.823610Z"},"content_sha256":"37961cb2d36696e4adfb56391ec9eb69423825d063cf6fef7a5615291740099b","schema_version":"1.0","event_id":"sha256:37961cb2d36696e4adfb56391ec9eb69423825d063cf6fef7a5615291740099b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BKGPZG6XQZLZE25PHPIGXEOLZU/bundle.json","state_url":"https://pith.science/pith/BKGPZG6XQZLZE25PHPIGXEOLZU/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BKGPZG6XQZLZE25PHPIGXEOLZU/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":"2026-08-10T01:22:51Z","links":{"resolver":"https://pith.science/pith/BKGPZG6XQZLZE25PHPIGXEOLZU","bundle":"https://pith.science/pith/BKGPZG6XQZLZE25PHPIGXEOLZU/bundle.json","state":"https://pith.science/pith/BKGPZG6XQZLZE25PHPIGXEOLZU/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BKGPZG6XQZLZE25PHPIGXEOLZU/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:BKGPZG6XQZLZE25PHPIGXEOLZU","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":"567daf31865dcb0ac0d685472a9687b0f0bf65f69397be48d22c0e5e9d60731c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-01T17:37:31Z","title_canon_sha256":"3d7678736fdc7023e8ff56796aef0736c78e2c4e84845c8cb481abbc9a16ef91"},"schema_version":"1.0","source":{"id":"2310.00741","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.00741","created_at":"2026-07-05T07:17:32Z"},{"alias_kind":"arxiv_version","alias_value":"2310.00741v2","created_at":"2026-07-05T07:17:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.00741","created_at":"2026-07-05T07:17:32Z"},{"alias_kind":"pith_short_12","alias_value":"BKGPZG6XQZLZ","created_at":"2026-07-05T07:17:32Z"},{"alias_kind":"pith_short_16","alias_value":"BKGPZG6XQZLZE25P","created_at":"2026-07-05T07:17:32Z"},{"alias_kind":"pith_short_8","alias_value":"BKGPZG6X","created_at":"2026-07-05T07:17:32Z"}],"graph_snapshots":[{"event_id":"sha256:37961cb2d36696e4adfb56391ec9eb69423825d063cf6fef7a5615291740099b","target":"graph","created_at":"2026-07-05T07:17:32Z","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/2310.00741/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Assessing factuality of text generated by large language models (LLMs) is an emerging yet crucial research area, aimed at alerting users to potential errors and guiding the development of more reliable LLMs. Nonetheless, the evaluators assessing factuality necessitate suitable evaluation themselves to gauge progress and foster advancements. This direction remains under-explored, resulting in substantial impediments to the progress of factuality evaluators. To mitigate this issue, we introduce a benchmark for Factuality Evaluation of large Language Models, referred to as felm. In this benchmark","authors_text":"I-Chun Chern, Jinghan Zhang, Junxian He, Pengfei Liu, Shiqi Chen, Siyang Gao, Yiran Zhao","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-01T17:37:31Z","title":"FELM: Benchmarking Factuality Evaluation of Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.00741","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:fe7907b6d9e6348506987f3c0aea3336ed79a055376e875f96bd7a305bb130c4","target":"record","created_at":"2026-07-05T07:17:32Z","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":"567daf31865dcb0ac0d685472a9687b0f0bf65f69397be48d22c0e5e9d60731c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-01T17:37:31Z","title_canon_sha256":"3d7678736fdc7023e8ff56796aef0736c78e2c4e84845c8cb481abbc9a16ef91"},"schema_version":"1.0","source":{"id":"2310.00741","kind":"arxiv","version":2}},"canonical_sha256":"0a8cfc9bd78657926baf3bd06b91cbcd317b8275c05d2d8ce08746a26d06493e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0a8cfc9bd78657926baf3bd06b91cbcd317b8275c05d2d8ce08746a26d06493e","first_computed_at":"2026-07-05T07:17:32.001749Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:17:32.001749Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"uNq+wbWmOsDAz6JoKB7J0AhjC1Q/qgJK/JOuyUZzshRqaSxyQpaMRJ+lzbMN0nDUzgz4yXhlUlYnYj8SCrZ0Bg==","signature_status":"signed_v1","signed_at":"2026-07-05T07:17:32.002206Z","signed_message":"canonical_sha256_bytes"},"source_id":"2310.00741","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:fe7907b6d9e6348506987f3c0aea3336ed79a055376e875f96bd7a305bb130c4","sha256:37961cb2d36696e4adfb56391ec9eb69423825d063cf6fef7a5615291740099b"],"state_sha256":"eaedb2c80f3737664bc554a05a8eff7e28f47a18da02a3015b614d9c6eb5e427"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MW9ZkK8Keql5RNFasjXm4DwQDsIFEOvTwdhkVijLF4vI56N3pWHu8GUfqC5nAdAMD/mraatqEUFCGuXaiGxRAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T01:22:51.829123Z","bundle_sha256":"c7d0d1fd4d4fe355a8680291a0a16172e05116ed556b827546e7567e251faea8"}}