{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:PY65BKKQIJQ2R5MK7XOGXSFIQ6","short_pith_number":"pith:PY65BKKQ","schema_version":"1.0","canonical_sha256":"7e3dd0a9504261a8f58afddc6bc8a8879bab925a948b23b8d0e04ad5792eed8c","source":{"kind":"arxiv","id":"2309.10621","version":3},"attestation_state":"computed","paper":{"title":"Large language models can accurately predict searcher preferences","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.IR","authors_text":"Bhaskar Mitra, Nick Craswell, Paul Thomas, Seth Spielman","submitted_at":"2023-09-19T13:55:39Z","abstract_excerpt":"Relevance labels, which indicate whether a search result is valuable to a searcher, are key to evaluating and optimising search systems. The best way to capture the true preferences of users is to ask them for their careful feedback on which results would be useful, but this approach does not scale to produce a large number of labels. Getting relevance labels at scale is usually done with third-party labellers, who judge on behalf of the user, but there is a risk of low-quality data if the labeller doesn't understand user needs. To improve quality, one standard approach is to study real users "},"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":"2309.10621","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2023-09-19T13:55:39Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"aa0a3914628e44bed63f52f9a2592e3be3ebe24a3373bdca4d734e05a6f02967","abstract_canon_sha256":"e59748bba856ef2e072f2e91401a2ca706314d2d0b35b9e41a307d604eba51de"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:19:56.686483Z","signature_b64":"V0wMxgkLX8yJy0n995cNBcGBj8eQ5SFXVQjNU/7sxUGVsYp87ly39/aq8ISBbxBx2zOG/Cdpnmmq0HUckn4wAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7e3dd0a9504261a8f58afddc6bc8a8879bab925a948b23b8d0e04ad5792eed8c","last_reissued_at":"2026-07-05T08:19:56.685993Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:19:56.685993Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Large language models can accurately predict searcher preferences","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.IR","authors_text":"Bhaskar Mitra, Nick Craswell, Paul Thomas, Seth Spielman","submitted_at":"2023-09-19T13:55:39Z","abstract_excerpt":"Relevance labels, which indicate whether a search result is valuable to a searcher, are key to evaluating and optimising search systems. The best way to capture the true preferences of users is to ask them for their careful feedback on which results would be useful, but this approach does not scale to produce a large number of labels. Getting relevance labels at scale is usually done with third-party labellers, who judge on behalf of the user, but there is a risk of low-quality data if the labeller doesn't understand user needs. To improve quality, one standard approach is to study real users "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.10621","kind":"arxiv","version":3},"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/2309.10621/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":"2309.10621","created_at":"2026-07-05T08:19:56.686053+00:00"},{"alias_kind":"arxiv_version","alias_value":"2309.10621v3","created_at":"2026-07-05T08:19:56.686053+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.10621","created_at":"2026-07-05T08:19:56.686053+00:00"},{"alias_kind":"pith_short_12","alias_value":"PY65BKKQIJQ2","created_at":"2026-07-05T08:19:56.686053+00:00"},{"alias_kind":"pith_short_16","alias_value":"PY65BKKQIJQ2R5MK","created_at":"2026-07-05T08:19:56.686053+00:00"},{"alias_kind":"pith_short_8","alias_value":"PY65BKKQ","created_at":"2026-07-05T08:19:56.686053+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.30840","citing_title":"Contrastive Reflection for Iterative Prompt Optimization","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2604.25683","citing_title":"K-CARE: Knowledge-driven Symmetrical Contextual Anchoring and Analogical Prototype Reasoning for E-commerce Relevance","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PY65BKKQIJQ2R5MK7XOGXSFIQ6","json":"https://pith.science/pith/PY65BKKQIJQ2R5MK7XOGXSFIQ6.json","graph_json":"https://pith.science/api/pith-number/PY65BKKQIJQ2R5MK7XOGXSFIQ6/graph.json","events_json":"https://pith.science/api/pith-number/PY65BKKQIJQ2R5MK7XOGXSFIQ6/events.json","paper":"https://pith.science/paper/PY65BKKQ"},"agent_actions":{"view_html":"https://pith.science/pith/PY65BKKQIJQ2R5MK7XOGXSFIQ6","download_json":"https://pith.science/pith/PY65BKKQIJQ2R5MK7XOGXSFIQ6.json","view_paper":"https://pith.science/paper/PY65BKKQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2309.10621&json=true","fetch_graph":"https://pith.science/api/pith-number/PY65BKKQIJQ2R5MK7XOGXSFIQ6/graph.json","fetch_events":"https://pith.science/api/pith-number/PY65BKKQIJQ2R5MK7XOGXSFIQ6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PY65BKKQIJQ2R5MK7XOGXSFIQ6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PY65BKKQIJQ2R5MK7XOGXSFIQ6/action/storage_attestation","attest_author":"https://pith.science/pith/PY65BKKQIJQ2R5MK7XOGXSFIQ6/action/author_attestation","sign_citation":"https://pith.science/pith/PY65BKKQIJQ2R5MK7XOGXSFIQ6/action/citation_signature","submit_replication":"https://pith.science/pith/PY65BKKQIJQ2R5MK7XOGXSFIQ6/action/replication_record"}},"created_at":"2026-07-05T08:19:56.686053+00:00","updated_at":"2026-07-05T08:19:56.686053+00:00"}