{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:IAVOAHVFQB5YLU4A5HRSEJFVCQ","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":"4d30c365da1982d0adebf503cd7e29fb1def6bf08083105511bd379e5e71f654","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-07T08:25:46Z","title_canon_sha256":"3b8a03f09bef3c10686337e6c42500ebb8981c4e9561cd7cf1d58ed9be915c40"},"schema_version":"1.0","source":{"id":"2403.04307","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.04307","created_at":"2026-07-05T09:06:57Z"},{"alias_kind":"arxiv_version","alias_value":"2403.04307v3","created_at":"2026-07-05T09:06:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.04307","created_at":"2026-07-05T09:06:57Z"},{"alias_kind":"pith_short_12","alias_value":"IAVOAHVFQB5Y","created_at":"2026-07-05T09:06:57Z"},{"alias_kind":"pith_short_16","alias_value":"IAVOAHVFQB5YLU4A","created_at":"2026-07-05T09:06:57Z"},{"alias_kind":"pith_short_8","alias_value":"IAVOAHVF","created_at":"2026-07-05T09:06:57Z"}],"graph_snapshots":[{"event_id":"sha256:26b0265479735b8e968a12f8644623c1c8202f058ab03ff9091fbf3ab195499e","target":"graph","created_at":"2026-07-05T09:06:57Z","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/2403.04307/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Hallucinations pose a significant challenge to the reliability of large language models (LLMs) in critical domains. Recent benchmarks designed to assess LLM hallucinations within conventional NLP tasks, such as knowledge-intensive question answering (QA) and summarization, are insufficient for capturing the complexities of user-LLM interactions in dynamic, real-world settings. To address this gap, we introduce HaluEval-Wild, the first benchmark specifically designed to evaluate LLM hallucinations in the wild. We meticulously collect challenging (adversarially filtered by Alpaca) user queries f","authors_text":"Yiming Yang, Zhiqing Sun, Zhiying Zhu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-07T08:25:46Z","title":"HaluEval-Wild: Evaluating Hallucinations of Language Models in the Wild"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.04307","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:acdfed2fd8c2720c3824518813316abca5365f938a9834b0f195108172111367","target":"record","created_at":"2026-07-05T09:06:57Z","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":"4d30c365da1982d0adebf503cd7e29fb1def6bf08083105511bd379e5e71f654","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-07T08:25:46Z","title_canon_sha256":"3b8a03f09bef3c10686337e6c42500ebb8981c4e9561cd7cf1d58ed9be915c40"},"schema_version":"1.0","source":{"id":"2403.04307","kind":"arxiv","version":3}},"canonical_sha256":"402ae01ea5807b85d380e9e32224b514306a1a9219691ca2830d72d140736ba5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"402ae01ea5807b85d380e9e32224b514306a1a9219691ca2830d72d140736ba5","first_computed_at":"2026-07-05T09:06:57.786115Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:06:57.786115Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"YL1ue+vJi7UkF6ML7Mkei1t7y0EkvYRzGxHAc0vx3eRJqPImmetIJU8LzX3wFEoLrWkO6OmNHlh1R/w7e3KgAg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:06:57.786617Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.04307","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:acdfed2fd8c2720c3824518813316abca5365f938a9834b0f195108172111367","sha256:26b0265479735b8e968a12f8644623c1c8202f058ab03ff9091fbf3ab195499e"],"state_sha256":"ef7c41959a3964105f1c164c27758b457134537945e75b3c791291d5192b4d6f"}