{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:JCG5HHMN5KVZWKCA7GWBHY5ORU","short_pith_number":"pith:JCG5HHMN","schema_version":"1.0","canonical_sha256":"488dd39d8deaab9b2840f9ac13e3ae8d16aab2105dc493969274c012095f08ec","source":{"kind":"arxiv","id":"2304.09848","version":2},"attestation_state":"computed","paper":{"title":"Evaluating Verifiability in Generative Search Engines","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CL","authors_text":"Nelson F. Liu, Percy Liang, Tianyi Zhang","submitted_at":"2023-04-19T17:56:12Z","abstract_excerpt":"Generative search engines directly generate responses to user queries, along with in-line citations. A prerequisite trait of a trustworthy generative search engine is verifiability, i.e., systems should cite comprehensively (high citation recall; all statements are fully supported by citations) and accurately (high citation precision; every cite supports its associated statement). We conduct human evaluation to audit four popular generative search engines -- Bing Chat, NeevaAI, perplexity.ai, and YouChat -- across a diverse set of queries from a variety of sources (e.g., historical Google user"},"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":"2304.09848","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-04-19T17:56:12Z","cross_cats_sorted":["cs.IR"],"title_canon_sha256":"494bfbf4b1564192caf382ee4735a23a4273f261072cc7e97235391adda065e6","abstract_canon_sha256":"e0c1626c91b631917471112c6b84397733db0959481da4af1630e38999640398"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:04:14.680844Z","signature_b64":"nmJ39344QWBCGTGuE9bXlqzKkHRlKBaIqvahxZMRDAqjQnwiXbtbebgZjijbrg/hGZh325pQhO+KhZqV6bHHAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"488dd39d8deaab9b2840f9ac13e3ae8d16aab2105dc493969274c012095f08ec","last_reissued_at":"2026-07-05T07:04:14.680397Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:04:14.680397Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Evaluating Verifiability in Generative Search Engines","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CL","authors_text":"Nelson F. Liu, Percy Liang, Tianyi Zhang","submitted_at":"2023-04-19T17:56:12Z","abstract_excerpt":"Generative search engines directly generate responses to user queries, along with in-line citations. A prerequisite trait of a trustworthy generative search engine is verifiability, i.e., systems should cite comprehensively (high citation recall; all statements are fully supported by citations) and accurately (high citation precision; every cite supports its associated statement). We conduct human evaluation to audit four popular generative search engines -- Bing Chat, NeevaAI, perplexity.ai, and YouChat -- across a diverse set of queries from a variety of sources (e.g., historical Google user"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2304.09848","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/2304.09848/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":"2304.09848","created_at":"2026-07-05T07:04:14.680454+00:00"},{"alias_kind":"arxiv_version","alias_value":"2304.09848v2","created_at":"2026-07-05T07:04:14.680454+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2304.09848","created_at":"2026-07-05T07:04:14.680454+00:00"},{"alias_kind":"pith_short_12","alias_value":"JCG5HHMN5KVZ","created_at":"2026-07-05T07:04:14.680454+00:00"},{"alias_kind":"pith_short_16","alias_value":"JCG5HHMN5KVZWKCA","created_at":"2026-07-05T07:04:14.680454+00:00"},{"alias_kind":"pith_short_8","alias_value":"JCG5HHMN","created_at":"2026-07-05T07:04:14.680454+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.28358","citing_title":"How Do LLMs Cite? A Mechanistic Interpretation of Attribution in Retrieval-Augmented Generation","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23684","citing_title":"Synthetic Sources?: Auditing Generative Search Engine Citations for Evidence of AI-Generated Sources","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2401.18059","citing_title":"RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrieval","ref_index":133,"is_internal_anchor":false},{"citing_arxiv_id":"2310.11511","citing_title":"Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection","ref_index":123,"is_internal_anchor":false},{"citing_arxiv_id":"2305.06161","citing_title":"StarCoder: may the source be with you!","ref_index":277,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17843","citing_title":"Learning from AVA: Early Lessons from a Curated and Trustworthy Generative AI for Policy and Development Research","ref_index":69,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JCG5HHMN5KVZWKCA7GWBHY5ORU","json":"https://pith.science/pith/JCG5HHMN5KVZWKCA7GWBHY5ORU.json","graph_json":"https://pith.science/api/pith-number/JCG5HHMN5KVZWKCA7GWBHY5ORU/graph.json","events_json":"https://pith.science/api/pith-number/JCG5HHMN5KVZWKCA7GWBHY5ORU/events.json","paper":"https://pith.science/paper/JCG5HHMN"},"agent_actions":{"view_html":"https://pith.science/pith/JCG5HHMN5KVZWKCA7GWBHY5ORU","download_json":"https://pith.science/pith/JCG5HHMN5KVZWKCA7GWBHY5ORU.json","view_paper":"https://pith.science/paper/JCG5HHMN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2304.09848&json=true","fetch_graph":"https://pith.science/api/pith-number/JCG5HHMN5KVZWKCA7GWBHY5ORU/graph.json","fetch_events":"https://pith.science/api/pith-number/JCG5HHMN5KVZWKCA7GWBHY5ORU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JCG5HHMN5KVZWKCA7GWBHY5ORU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JCG5HHMN5KVZWKCA7GWBHY5ORU/action/storage_attestation","attest_author":"https://pith.science/pith/JCG5HHMN5KVZWKCA7GWBHY5ORU/action/author_attestation","sign_citation":"https://pith.science/pith/JCG5HHMN5KVZWKCA7GWBHY5ORU/action/citation_signature","submit_replication":"https://pith.science/pith/JCG5HHMN5KVZWKCA7GWBHY5ORU/action/replication_record"}},"created_at":"2026-07-05T07:04:14.680454+00:00","updated_at":"2026-07-05T07:04:14.680454+00:00"}