{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:7W5N6FZWA76DPAORYCXT6TZLH7","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":"9e3a0fe544ecf93e812eacb5478f5f368e2c9359e1d681262c71b1be91dbe9a9","cross_cats_sorted":["cs.AI","cs.CV","cs.IR","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-18T16:27:49Z","title_canon_sha256":"53369d1df7cac9ac1c1f3968895717d2e605ca8c4c0e4cc2930d111360ecca3a"},"schema_version":"1.0","source":{"id":"2310.12086","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.12086","created_at":"2026-07-05T08:23:12Z"},{"alias_kind":"arxiv_version","alias_value":"2310.12086v3","created_at":"2026-07-05T08:23:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.12086","created_at":"2026-07-05T08:23:12Z"},{"alias_kind":"pith_short_12","alias_value":"7W5N6FZWA76D","created_at":"2026-07-05T08:23:12Z"},{"alias_kind":"pith_short_16","alias_value":"7W5N6FZWA76DPAOR","created_at":"2026-07-05T08:23:12Z"},{"alias_kind":"pith_short_8","alias_value":"7W5N6FZW","created_at":"2026-07-05T08:23:12Z"}],"graph_snapshots":[{"event_id":"sha256:d522ca99666171303c470153149a8e50be3e89b2884acc2676a3893a9cd04887","target":"graph","created_at":"2026-07-05T08:23:12Z","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/2310.12086/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Despite their impressive generative capabilities, LLMs are hindered by fact-conflicting hallucinations in real-world applications. The accurate identification of hallucinations in texts generated by LLMs, especially in complex inferential scenarios, is a relatively unexplored area. To address this gap, we present FactCHD, a dedicated benchmark designed for the detection of fact-conflicting hallucinations from LLMs. FactCHD features a diverse dataset that spans various factuality patterns, including vanilla, multi-hop, comparison, and set operation. A distinctive element of FactCHD is its integ","authors_text":"Chengfei Lv, Chenxi Wang, Dan Zhang, Duanzheng Song, Fei Huang, Honghao Gui, Huajun Chen, Ningyu Zhang, Xiang Chen, Yong Jiang","cross_cats":["cs.AI","cs.CV","cs.IR","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-18T16:27:49Z","title":"FactCHD: Benchmarking Fact-Conflicting Hallucination Detection"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.12086","kind":"arxiv","version":3},"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:a77093df0e285e9d7933930cf8ae11c41b4e3e776d533d30fe8a1217abb70b41","target":"record","created_at":"2026-07-05T08:23:12Z","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":"9e3a0fe544ecf93e812eacb5478f5f368e2c9359e1d681262c71b1be91dbe9a9","cross_cats_sorted":["cs.AI","cs.CV","cs.IR","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-18T16:27:49Z","title_canon_sha256":"53369d1df7cac9ac1c1f3968895717d2e605ca8c4c0e4cc2930d111360ecca3a"},"schema_version":"1.0","source":{"id":"2310.12086","kind":"arxiv","version":3}},"canonical_sha256":"fdbadf173607fc3781d1c0af3f4f2b3ff9817d3921d4976ce2c53035c3992b8b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"fdbadf173607fc3781d1c0af3f4f2b3ff9817d3921d4976ce2c53035c3992b8b","first_computed_at":"2026-07-05T08:23:12.995838Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:23:12.995838Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"xbnHkhaYi5vXU7GvlFikF57CDWQxL2ouH/X/8RweRG5HZ/U9GGsTtlyyoAgZ4QjFeBkkieUU6KnCNzb8d3B0CA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:23:12.996280Z","signed_message":"canonical_sha256_bytes"},"source_id":"2310.12086","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a77093df0e285e9d7933930cf8ae11c41b4e3e776d533d30fe8a1217abb70b41","sha256:d522ca99666171303c470153149a8e50be3e89b2884acc2676a3893a9cd04887"],"state_sha256":"57e347621c13daa6eac2a366ce6946f01a83cdebfc94bd00d8af5bcb966ecac7"}