{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:MZFUZAAV5IVIV7SUH4UHX7QCCD","short_pith_number":"pith:MZFUZAAV","schema_version":"1.0","canonical_sha256":"664b4c8015ea2a8afe543f287bfe0210e1df91129bf0412ee0063468f85c2139","source":{"kind":"arxiv","id":"2502.15845","version":2},"attestation_state":"computed","paper":{"title":"Verify when Uncertain: Beyond Self-Consistency in Black Box Hallucination Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Baharan Mirzasoleiman, Kristjan Greenewald, Yihao Xue, Youssef Mroueh","submitted_at":"2025-02-20T21:06:08Z","abstract_excerpt":"Large Language Models (LLMs) often hallucinate, limiting their reliability in sensitive applications. In black-box settings, several self-consistency-based techniques have been proposed for hallucination detection. We empirically show that these methods perform nearly as well as a supervised (black-box) oracle, leaving limited room for further gains within this paradigm. To address this limitation, we explore cross-model consistency checking between the target model and an additional verifier LLM. With this extra information, we observe improved oracle performance compared to purely self-consi"},"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":"2502.15845","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-20T21:06:08Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"0d974ee802519e0f93832ac232cfc9c44ab0e00c531ccc0b7da4d3f9c8ab69b2","abstract_canon_sha256":"048c56b92caffca42f4998f13b3a17f3f2196cad32098bdeaba5cff46d055fab"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-01T00:17:07.913534Z","signature_b64":"RkzjPYcCypu9QKO1Zl52f9fOOd4sAEv5tgtY4OhRttiZSvpdA+nZIBawzqaECfQX2GVr6mUv5yOHOyFRnyTlAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"664b4c8015ea2a8afe543f287bfe0210e1df91129bf0412ee0063468f85c2139","last_reissued_at":"2026-07-01T00:17:07.913000Z","signature_status":"signed_v1","first_computed_at":"2026-07-01T00:17:07.913000Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Verify when Uncertain: Beyond Self-Consistency in Black Box Hallucination Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Baharan Mirzasoleiman, Kristjan Greenewald, Yihao Xue, Youssef Mroueh","submitted_at":"2025-02-20T21:06:08Z","abstract_excerpt":"Large Language Models (LLMs) often hallucinate, limiting their reliability in sensitive applications. In black-box settings, several self-consistency-based techniques have been proposed for hallucination detection. We empirically show that these methods perform nearly as well as a supervised (black-box) oracle, leaving limited room for further gains within this paradigm. To address this limitation, we explore cross-model consistency checking between the target model and an additional verifier LLM. With this extra information, we observe improved oracle performance compared to purely self-consi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.15845","kind":"arxiv","version":2},"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/2502.15845/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":"2502.15845","created_at":"2026-07-01T00:17:07.913078+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.15845v2","created_at":"2026-07-01T00:17:07.913078+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.15845","created_at":"2026-07-01T00:17:07.913078+00:00"},{"alias_kind":"pith_short_12","alias_value":"MZFUZAAV5IVI","created_at":"2026-07-01T00:17:07.913078+00:00"},{"alias_kind":"pith_short_16","alias_value":"MZFUZAAV5IVIV7SU","created_at":"2026-07-01T00:17:07.913078+00:00"},{"alias_kind":"pith_short_8","alias_value":"MZFUZAAV","created_at":"2026-07-01T00:17:07.913078+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":6,"sample":[{"citing_arxiv_id":"2604.02784","citing_title":"EnsemHalDet: Robust VLM Hallucination Detection via Ensemble of Internal State Detectors","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2604.02784","citing_title":"EnsemHalDet: Robust VLM Hallucination Detection via Ensemble of Internal State Detectors","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2605.05103","citing_title":"Text Corpora as Concept Fields: Black-Box Hallucination and Novelty Measurement","ref_index":5,"is_internal_anchor":true},{"citing_arxiv_id":"2605.05103","citing_title":"Text Corpora as Concept Fields: Black-Box Hallucination and Novelty Measurement","ref_index":5,"is_internal_anchor":true},{"citing_arxiv_id":"2604.19162","citing_title":"Mind the Unseen Mass: Unmasking LLM Hallucinations via Soft-Hybrid Alphabet Estimation","ref_index":34,"is_internal_anchor":true},{"citing_arxiv_id":"2604.17112","citing_title":"Complementing Self-Consistency with Cross-Model Disagreement for Uncertainty Quantification","ref_index":48,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MZFUZAAV5IVIV7SUH4UHX7QCCD","json":"https://pith.science/pith/MZFUZAAV5IVIV7SUH4UHX7QCCD.json","graph_json":"https://pith.science/api/pith-number/MZFUZAAV5IVIV7SUH4UHX7QCCD/graph.json","events_json":"https://pith.science/api/pith-number/MZFUZAAV5IVIV7SUH4UHX7QCCD/events.json","paper":"https://pith.science/paper/MZFUZAAV"},"agent_actions":{"view_html":"https://pith.science/pith/MZFUZAAV5IVIV7SUH4UHX7QCCD","download_json":"https://pith.science/pith/MZFUZAAV5IVIV7SUH4UHX7QCCD.json","view_paper":"https://pith.science/paper/MZFUZAAV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.15845&json=true","fetch_graph":"https://pith.science/api/pith-number/MZFUZAAV5IVIV7SUH4UHX7QCCD/graph.json","fetch_events":"https://pith.science/api/pith-number/MZFUZAAV5IVIV7SUH4UHX7QCCD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MZFUZAAV5IVIV7SUH4UHX7QCCD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MZFUZAAV5IVIV7SUH4UHX7QCCD/action/storage_attestation","attest_author":"https://pith.science/pith/MZFUZAAV5IVIV7SUH4UHX7QCCD/action/author_attestation","sign_citation":"https://pith.science/pith/MZFUZAAV5IVIV7SUH4UHX7QCCD/action/citation_signature","submit_replication":"https://pith.science/pith/MZFUZAAV5IVIV7SUH4UHX7QCCD/action/replication_record"}},"created_at":"2026-07-01T00:17:07.913078+00:00","updated_at":"2026-07-01T00:17:07.913078+00:00"}