{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:NP5MRU26QWLG676M62V2VJCP35","short_pith_number":"pith:NP5MRU26","schema_version":"1.0","canonical_sha256":"6bfac8d35e85966f7fccf6abaaa44fdf67a6fcf5b33091f144f2a133388b523b","source":{"kind":"arxiv","id":"2302.02657","version":1},"attestation_state":"computed","paper":{"title":"Divide and Conquer: Towards Better Embedding-based Retrieval for Recommender Systems From a Multi-task Perspective","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Biao Li, Kun Gai, Peng Jiang, Weijie Ding, Xue Dong, Yuan Zhang","submitted_at":"2023-02-06T09:52:12Z","abstract_excerpt":"Embedding-based retrieval (EBR) methods are widely used in modern recommender systems thanks to its simplicity and effectiveness. However, along the journey of deploying and iterating on EBR in production, we still identify some fundamental issues in existing methods. First, when dealing with large corpus of candidate items, EBR models often have difficulties in balancing the performance on distinguishing highly relevant items (positives) from both irrelevant ones (easy negatives) and from somewhat related yet not competitive ones (hard negatives). Also, we have little control in the diversity"},"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":"2302.02657","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2023-02-06T09:52:12Z","cross_cats_sorted":[],"title_canon_sha256":"1e3e21f53814baac7dc1615fa9bee9f7cdc8202e1ccb82ddc9d05411bca5fe07","abstract_canon_sha256":"7354b92b41da4e7d1c15458d6cfe3cacc3a8a87d44f17e4972e6e9e319b5dd4d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:39:05.598660Z","signature_b64":"tj3sFYC/pramCMFdcPyd2QKccU+8zGBN5A7RUy+Tv0tm9z8RGxsi+vSZ/UbEBC3mcNmiw8wlL+dRHJhzCUROAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6bfac8d35e85966f7fccf6abaaa44fdf67a6fcf5b33091f144f2a133388b523b","last_reissued_at":"2026-07-05T05:39:05.598231Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:39:05.598231Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Divide and Conquer: Towards Better Embedding-based Retrieval for Recommender Systems From a Multi-task Perspective","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Biao Li, Kun Gai, Peng Jiang, Weijie Ding, Xue Dong, Yuan Zhang","submitted_at":"2023-02-06T09:52:12Z","abstract_excerpt":"Embedding-based retrieval (EBR) methods are widely used in modern recommender systems thanks to its simplicity and effectiveness. However, along the journey of deploying and iterating on EBR in production, we still identify some fundamental issues in existing methods. First, when dealing with large corpus of candidate items, EBR models often have difficulties in balancing the performance on distinguishing highly relevant items (positives) from both irrelevant ones (easy negatives) and from somewhat related yet not competitive ones (hard negatives). Also, we have little control in the diversity"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.02657","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/2302.02657/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":"2302.02657","created_at":"2026-07-05T05:39:05.598292+00:00"},{"alias_kind":"arxiv_version","alias_value":"2302.02657v1","created_at":"2026-07-05T05:39:05.598292+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.02657","created_at":"2026-07-05T05:39:05.598292+00:00"},{"alias_kind":"pith_short_12","alias_value":"NP5MRU26QWLG","created_at":"2026-07-05T05:39:05.598292+00:00"},{"alias_kind":"pith_short_16","alias_value":"NP5MRU26QWLG676M","created_at":"2026-07-05T05:39:05.598292+00:00"},{"alias_kind":"pith_short_8","alias_value":"NP5MRU26","created_at":"2026-07-05T05:39:05.598292+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/NP5MRU26QWLG676M62V2VJCP35","json":"https://pith.science/pith/NP5MRU26QWLG676M62V2VJCP35.json","graph_json":"https://pith.science/api/pith-number/NP5MRU26QWLG676M62V2VJCP35/graph.json","events_json":"https://pith.science/api/pith-number/NP5MRU26QWLG676M62V2VJCP35/events.json","paper":"https://pith.science/paper/NP5MRU26"},"agent_actions":{"view_html":"https://pith.science/pith/NP5MRU26QWLG676M62V2VJCP35","download_json":"https://pith.science/pith/NP5MRU26QWLG676M62V2VJCP35.json","view_paper":"https://pith.science/paper/NP5MRU26","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2302.02657&json=true","fetch_graph":"https://pith.science/api/pith-number/NP5MRU26QWLG676M62V2VJCP35/graph.json","fetch_events":"https://pith.science/api/pith-number/NP5MRU26QWLG676M62V2VJCP35/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NP5MRU26QWLG676M62V2VJCP35/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NP5MRU26QWLG676M62V2VJCP35/action/storage_attestation","attest_author":"https://pith.science/pith/NP5MRU26QWLG676M62V2VJCP35/action/author_attestation","sign_citation":"https://pith.science/pith/NP5MRU26QWLG676M62V2VJCP35/action/citation_signature","submit_replication":"https://pith.science/pith/NP5MRU26QWLG676M62V2VJCP35/action/replication_record"}},"created_at":"2026-07-05T05:39:05.598292+00:00","updated_at":"2026-07-05T05:39:05.598292+00:00"}