{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:MNCRD45AEMJFIGJCNTILMPEEN6","short_pith_number":"pith:MNCRD45A","schema_version":"1.0","canonical_sha256":"634511f3a023125419226cd0b63c846fbd14a395f851db485369822450e50a16","source":{"kind":"arxiv","id":"2205.06205","version":3},"attestation_state":"computed","paper":{"title":"kNN-Embed: Locally Smoothed Embedding Mixtures For Multi-interest Candidate Retrieval","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.IR","authors_text":"Ahmed El-Kishky, Aria Haghighi, Frank Portman, Kenny Leung, Thomas Markovich, Ying Xiao","submitted_at":"2022-05-12T16:42:24Z","abstract_excerpt":"Candidate retrieval is the first stage in recommendation systems, where a light-weight system is used to retrieve potentially relevant items for an input user. These candidate items are then ranked and pruned in later stages of recommender systems using a more complex ranking model. As the top of the recommendation funnel, it is important to retrieve a high-recall candidate set to feed into downstream ranking models. A common approach is to leverage approximate nearest neighbor (ANN) search from a single dense query embedding; however, this approach this can yield a low-diversity result set wi"},"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":"2205.06205","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2022-05-12T16:42:24Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"d489b1c5e1bc10f4c0ab83663305765386367512bceab6268de67532e8897a36","abstract_canon_sha256":"e1bf7e6d50f9c62254ccf3c250f87379212845a3a4c887cf474f750bdf17a297"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:37:49.162067Z","signature_b64":"gr9VeoI7kg3tzoGvqRP6KvblY4MnQRHDWD9MMrBDzprx0idq7kMeycAexXENp1MZSdLudr8tkG4uCTu336RRCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"634511f3a023125419226cd0b63c846fbd14a395f851db485369822450e50a16","last_reissued_at":"2026-07-05T06:37:49.161552Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:37:49.161552Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"kNN-Embed: Locally Smoothed Embedding Mixtures For Multi-interest Candidate Retrieval","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.IR","authors_text":"Ahmed El-Kishky, Aria Haghighi, Frank Portman, Kenny Leung, Thomas Markovich, Ying Xiao","submitted_at":"2022-05-12T16:42:24Z","abstract_excerpt":"Candidate retrieval is the first stage in recommendation systems, where a light-weight system is used to retrieve potentially relevant items for an input user. These candidate items are then ranked and pruned in later stages of recommender systems using a more complex ranking model. As the top of the recommendation funnel, it is important to retrieve a high-recall candidate set to feed into downstream ranking models. A common approach is to leverage approximate nearest neighbor (ANN) search from a single dense query embedding; however, this approach this can yield a low-diversity result set wi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.06205","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/2205.06205/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":"2205.06205","created_at":"2026-07-05T06:37:49.161610+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.06205v3","created_at":"2026-07-05T06:37:49.161610+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.06205","created_at":"2026-07-05T06:37:49.161610+00:00"},{"alias_kind":"pith_short_12","alias_value":"MNCRD45AEMJF","created_at":"2026-07-05T06:37:49.161610+00:00"},{"alias_kind":"pith_short_16","alias_value":"MNCRD45AEMJFIGJC","created_at":"2026-07-05T06:37:49.161610+00:00"},{"alias_kind":"pith_short_8","alias_value":"MNCRD45A","created_at":"2026-07-05T06:37:49.161610+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MNCRD45AEMJFIGJCNTILMPEEN6","json":"https://pith.science/pith/MNCRD45AEMJFIGJCNTILMPEEN6.json","graph_json":"https://pith.science/api/pith-number/MNCRD45AEMJFIGJCNTILMPEEN6/graph.json","events_json":"https://pith.science/api/pith-number/MNCRD45AEMJFIGJCNTILMPEEN6/events.json","paper":"https://pith.science/paper/MNCRD45A"},"agent_actions":{"view_html":"https://pith.science/pith/MNCRD45AEMJFIGJCNTILMPEEN6","download_json":"https://pith.science/pith/MNCRD45AEMJFIGJCNTILMPEEN6.json","view_paper":"https://pith.science/paper/MNCRD45A","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.06205&json=true","fetch_graph":"https://pith.science/api/pith-number/MNCRD45AEMJFIGJCNTILMPEEN6/graph.json","fetch_events":"https://pith.science/api/pith-number/MNCRD45AEMJFIGJCNTILMPEEN6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MNCRD45AEMJFIGJCNTILMPEEN6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MNCRD45AEMJFIGJCNTILMPEEN6/action/storage_attestation","attest_author":"https://pith.science/pith/MNCRD45AEMJFIGJCNTILMPEEN6/action/author_attestation","sign_citation":"https://pith.science/pith/MNCRD45AEMJFIGJCNTILMPEEN6/action/citation_signature","submit_replication":"https://pith.science/pith/MNCRD45AEMJFIGJCNTILMPEEN6/action/replication_record"}},"created_at":"2026-07-05T06:37:49.161610+00:00","updated_at":"2026-07-05T06:37:49.161610+00:00"}