{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:22ABGJOR47GY5DQ544C7N5I5AP","short_pith_number":"pith:22ABGJOR","schema_version":"1.0","canonical_sha256":"d6801325d1e7cd8e8e1de705f6f51d03cf209184ee9732d1490b8c84c5388b0f","source":{"kind":"arxiv","id":"2407.17468","version":1},"attestation_state":"computed","paper":{"title":"WildHallucinations: Evaluating Long-form Factuality in LLMs with Real-World Entity Queries","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Abhilasha Ravichander, Benjamin Newman, Claire Cardie, Khyathi Chandu, Liwei Jiang, Ronan Le Bras, Tanya Goyal, Wenting Zhao, Yejin Choi, Yuntian Deng, Yu Ying Chiu","submitted_at":"2024-07-24T17:59:05Z","abstract_excerpt":"While hallucinations of large language models (LLMs) prevail as a major challenge, existing evaluation benchmarks on factuality do not cover the diverse domains of knowledge that the real-world users of LLMs seek information about. To bridge this gap, we introduce WildHallucinations, a benchmark that evaluates factuality. It does so by prompting LLMs to generate information about entities mined from user-chatbot conversations in the wild. These generations are then automatically fact-checked against a systematically curated knowledge source collected from web search. Notably, half of these rea"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2407.17468","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-24T17:59:05Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"73fa7dc1142c14c46879e3a1745145be2e46e4ce72183dd9c65330a8de8f39fd","abstract_canon_sha256":"bdc7ede44c0452bee59c77245df24f88dc9ca463c76d490d14c301f7b46657d1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:48:08.963444Z","signature_b64":"Rs3fgqtiVn7dHP/csaFIeJ7NUMH0IaAEemCkpnGKWgjdkkhwADBIey/nH2GPgeJSQNuWQjJLmNbC4VUvMEDADA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d6801325d1e7cd8e8e1de705f6f51d03cf209184ee9732d1490b8c84c5388b0f","last_reissued_at":"2026-07-05T08:48:08.962931Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:48:08.962931Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"WildHallucinations: Evaluating Long-form Factuality in LLMs with Real-World Entity Queries","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Abhilasha Ravichander, Benjamin Newman, Claire Cardie, Khyathi Chandu, Liwei Jiang, Ronan Le Bras, Tanya Goyal, Wenting Zhao, Yejin Choi, Yuntian Deng, Yu Ying Chiu","submitted_at":"2024-07-24T17:59:05Z","abstract_excerpt":"While hallucinations of large language models (LLMs) prevail as a major challenge, existing evaluation benchmarks on factuality do not cover the diverse domains of knowledge that the real-world users of LLMs seek information about. To bridge this gap, we introduce WildHallucinations, a benchmark that evaluates factuality. It does so by prompting LLMs to generate information about entities mined from user-chatbot conversations in the wild. These generations are then automatically fact-checked against a systematically curated knowledge source collected from web search. Notably, half of these rea"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.17468","kind":"arxiv","version":1},"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/2407.17468/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2407.17468","created_at":"2026-07-05T08:48:08.963003+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.17468v1","created_at":"2026-07-05T08:48:08.963003+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.17468","created_at":"2026-07-05T08:48:08.963003+00:00"},{"alias_kind":"pith_short_12","alias_value":"22ABGJOR47GY","created_at":"2026-07-05T08:48:08.963003+00:00"},{"alias_kind":"pith_short_16","alias_value":"22ABGJOR47GY5DQ5","created_at":"2026-07-05T08:48:08.963003+00:00"},{"alias_kind":"pith_short_8","alias_value":"22ABGJOR","created_at":"2026-07-05T08:48:08.963003+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21595","citing_title":"Per-Entity Bias Mapping for AI Visibility: Why Brand Mentions Require Entity-Specific Calibration","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05054","citing_title":"Boosting Self-Consistency with Ranking","ref_index":267,"is_internal_anchor":false},{"citing_arxiv_id":"2505.23912","citing_title":"LoVeC: Reinforcement Learning for Better Verbalized Confidence in Long-Form Generations","ref_index":54,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00737","citing_title":"To Call or Not to Call: A Framework to Assess and Optimize LLM Tool Calling","ref_index":54,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/22ABGJOR47GY5DQ544C7N5I5AP","json":"https://pith.science/pith/22ABGJOR47GY5DQ544C7N5I5AP.json","graph_json":"https://pith.science/api/pith-number/22ABGJOR47GY5DQ544C7N5I5AP/graph.json","events_json":"https://pith.science/api/pith-number/22ABGJOR47GY5DQ544C7N5I5AP/events.json","paper":"https://pith.science/paper/22ABGJOR"},"agent_actions":{"view_html":"https://pith.science/pith/22ABGJOR47GY5DQ544C7N5I5AP","download_json":"https://pith.science/pith/22ABGJOR47GY5DQ544C7N5I5AP.json","view_paper":"https://pith.science/paper/22ABGJOR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.17468&json=true","fetch_graph":"https://pith.science/api/pith-number/22ABGJOR47GY5DQ544C7N5I5AP/graph.json","fetch_events":"https://pith.science/api/pith-number/22ABGJOR47GY5DQ544C7N5I5AP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/22ABGJOR47GY5DQ544C7N5I5AP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/22ABGJOR47GY5DQ544C7N5I5AP/action/storage_attestation","attest_author":"https://pith.science/pith/22ABGJOR47GY5DQ544C7N5I5AP/action/author_attestation","sign_citation":"https://pith.science/pith/22ABGJOR47GY5DQ544C7N5I5AP/action/citation_signature","submit_replication":"https://pith.science/pith/22ABGJOR47GY5DQ544C7N5I5AP/action/replication_record"}},"created_at":"2026-07-05T08:48:08.963003+00:00","updated_at":"2026-07-05T08:48:08.963003+00:00"}