{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:FHVBIEZQM6STQO7KATAF5PWMHJ","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":"501c5241f56b1498ea17fe4c4bc44d302a54c85526b13ae54e44c10dcdabd999","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-24T13:40:27Z","title_canon_sha256":"4887496cc92eb8b19aa488da5d2728ced47bd0c69fde59930ca40709d78e3f48"},"schema_version":"1.0","source":{"id":"2504.17550","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.17550","created_at":"2026-07-05T10:53:30Z"},{"alias_kind":"arxiv_version","alias_value":"2504.17550v1","created_at":"2026-07-05T10:53:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.17550","created_at":"2026-07-05T10:53:30Z"},{"alias_kind":"pith_short_12","alias_value":"FHVBIEZQM6ST","created_at":"2026-07-05T10:53:30Z"},{"alias_kind":"pith_short_16","alias_value":"FHVBIEZQM6STQO7K","created_at":"2026-07-05T10:53:30Z"},{"alias_kind":"pith_short_8","alias_value":"FHVBIEZQ","created_at":"2026-07-05T10:53:30Z"}],"graph_snapshots":[{"event_id":"sha256:90d8c5a836eea0d47b3e56b7918139638d7a0448e021143745fba2a90ee8e79d","target":"graph","created_at":"2026-07-05T10:53:30Z","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/2504.17550/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) often generate responses that deviate from user input or training data, a phenomenon known as \"hallucination.\" These hallucinations undermine user trust and hinder the adoption of generative AI systems. Addressing hallucinations is essential for the advancement of LLMs. This paper introduces a comprehensive hallucination benchmark, incorporating both new extrinsic and existing intrinsic evaluation tasks, built upon clear taxonomy of hallucination. A major challenge in benchmarking hallucinations is the lack of a unified framework due to inconsistent definitions and","authors_text":"Alan Schelten, Anthony Hartshorn, Cheng Zhang, Nicola Cancedda, Pascale Fung, Tara Fowler, Yejin Bang, Ziwei Ji","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-24T13:40:27Z","title":"HalluLens: LLM Hallucination Benchmark"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.17550","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":"sha256:3245a15981860980c30cf9c535231afd1a43fa2cbd6dba88b3c880a0f6fb0ec2","target":"record","created_at":"2026-07-05T10:53:30Z","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":"501c5241f56b1498ea17fe4c4bc44d302a54c85526b13ae54e44c10dcdabd999","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-24T13:40:27Z","title_canon_sha256":"4887496cc92eb8b19aa488da5d2728ced47bd0c69fde59930ca40709d78e3f48"},"schema_version":"1.0","source":{"id":"2504.17550","kind":"arxiv","version":1}},"canonical_sha256":"29ea14133067a5383bea04c05ebecc3a403010654d5b18d13ce7ab8ffefe36b3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"29ea14133067a5383bea04c05ebecc3a403010654d5b18d13ce7ab8ffefe36b3","first_computed_at":"2026-07-05T10:53:30.746275Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:53:30.746275Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9iYgnbF8ojn22TJDz+Mhp7YhIC8/O0Jqty7PrxuJsXNoglUkEXg1z8tGteeDfp2smRW5kv21GjWuxn7st8MvBw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:53:30.746867Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.17550","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3245a15981860980c30cf9c535231afd1a43fa2cbd6dba88b3c880a0f6fb0ec2","sha256:90d8c5a836eea0d47b3e56b7918139638d7a0448e021143745fba2a90ee8e79d"],"state_sha256":"c46df5ef8ba98bbb5603111d584dc492f1491f63d184d39f17359956603eac3d"}