{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:RUNH5PXJRNLNOA5R6JQ3H6OYWG","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":"8297249c1892ba770cf615dad5dae7c815448780302b513354a47064342bd685","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-25T07:44:22Z","title_canon_sha256":"ba9caa13c09373a4282ec633fd23b17790881068355a5daf50ec797848179aff"},"schema_version":"1.0","source":{"id":"2502.17924","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.17924","created_at":"2026-07-05T10:22:14Z"},{"alias_kind":"arxiv_version","alias_value":"2502.17924v2","created_at":"2026-07-05T10:22:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.17924","created_at":"2026-07-05T10:22:14Z"},{"alias_kind":"pith_short_12","alias_value":"RUNH5PXJRNLN","created_at":"2026-07-05T10:22:14Z"},{"alias_kind":"pith_short_16","alias_value":"RUNH5PXJRNLNOA5R","created_at":"2026-07-05T10:22:14Z"},{"alias_kind":"pith_short_8","alias_value":"RUNH5PXJ","created_at":"2026-07-05T10:22:14Z"}],"graph_snapshots":[{"event_id":"sha256:9fd06cf8a8e81aab6e5c40a11c75042fb7e28f8a091fc55cdfed7f8ff6fcb34b","target":"graph","created_at":"2026-07-05T10:22:14Z","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/2502.17924/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) have significantly advanced the fact-checking studies. However, existing automated fact-checking evaluation methods rely on static datasets and classification metrics, which fail to automatically evaluate the justification production and uncover the nuanced limitations of LLMs in fact-checking. In this work, we introduce FACT-AUDIT, an agent-driven framework that adaptively and dynamically assesses LLMs' fact-checking capabilities. Leveraging importance sampling principles and multi-agent collaboration, FACT-AUDIT generates adaptive and scalable datasets, performs ","authors_text":"Hongzhan Lin, Jing Ma, See-kiong Ng, Tat-Seng Chua, Wenxuan Zhang, Yang Deng, Yuxuan Gu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-25T07:44:22Z","title":"FACT-AUDIT: An Adaptive Multi-Agent Framework for Dynamic Fact-Checking Evaluation of Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.17924","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":"sha256:801d25964c063e6689e124123a37507f43fd4eb4a28ae862cd06281cf3bc2f29","target":"record","created_at":"2026-07-05T10:22:14Z","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":"8297249c1892ba770cf615dad5dae7c815448780302b513354a47064342bd685","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-25T07:44:22Z","title_canon_sha256":"ba9caa13c09373a4282ec633fd23b17790881068355a5daf50ec797848179aff"},"schema_version":"1.0","source":{"id":"2502.17924","kind":"arxiv","version":2}},"canonical_sha256":"8d1a7ebee98b56d703b1f261b3f9d8b18af6e57204ef192b24b3484ea565e165","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8d1a7ebee98b56d703b1f261b3f9d8b18af6e57204ef192b24b3484ea565e165","first_computed_at":"2026-07-05T10:22:14.528026Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:22:14.528026Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"zOWmtrM1DQd2gkxpZ6zNp6+RJuyhYpyM41qz0sY+k5jfXvdWmMFCWahmmU0YMWC+htnpowFCOILd7E9y1VGuAw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:22:14.528659Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.17924","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:801d25964c063e6689e124123a37507f43fd4eb4a28ae862cd06281cf3bc2f29","sha256:9fd06cf8a8e81aab6e5c40a11c75042fb7e28f8a091fc55cdfed7f8ff6fcb34b"],"state_sha256":"e63cf446fd4df57067acf32d161b8671a0e63c0877ce62210cb7392e0e95f861"}