{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:7OMXF6P4JLYNAVHEKVI7UYOGDT","short_pith_number":"pith:7OMXF6P4","schema_version":"1.0","canonical_sha256":"fb9972f9fc4af0d054e45551fa61c61ce8210e7aeba48bbba8d1f33048568c45","source":{"kind":"arxiv","id":"2505.11318","version":2},"attestation_state":"computed","paper":{"title":"On the Role of Weight Decay in Collaborative Filtering: A Popularity Perspective","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.IR","authors_text":"Danai Koutra, Donald Loveland, Mingxuan Ju, Neil Shah, Tong Zhao","submitted_at":"2025-05-16T14:41:57Z","abstract_excerpt":"Collaborative filtering (CF) enables large-scale recommendation systems by encoding information from historical user-item interactions into dense ID-embedding tables. However, as embedding tables grow, closed-form solutions become impractical, often necessitating the use of mini-batch gradient descent for training. Despite extensive work on designing loss functions to train CF models, we argue that one core component of these pipelines is heavily overlooked: weight decay. Attaining high-performing models typically requires careful tuning of weight decay, regardless of loss, yet its necessity i"},"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":"2505.11318","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2025-05-16T14:41:57Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"0d8ebd0b76d6aca4412ddf296fdf9c7119803f6cbe26e50b772b9df9068a3324","abstract_canon_sha256":"f8b93b11a70e8363417fefe89dde74d7ecd387a783d239f7c3dcdde9606b26ba"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:02:46.240085Z","signature_b64":"g/+bXaiAqPDvvILWa3AYWEi18pcnTGvWiZnT11CERL7zLaiQhlitU+eCq4s+vgr/lsQKznFTpMWIjInPoZ+qCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fb9972f9fc4af0d054e45551fa61c61ce8210e7aeba48bbba8d1f33048568c45","last_reissued_at":"2026-07-05T12:02:46.239580Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:02:46.239580Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On the Role of Weight Decay in Collaborative Filtering: A Popularity Perspective","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.IR","authors_text":"Danai Koutra, Donald Loveland, Mingxuan Ju, Neil Shah, Tong Zhao","submitted_at":"2025-05-16T14:41:57Z","abstract_excerpt":"Collaborative filtering (CF) enables large-scale recommendation systems by encoding information from historical user-item interactions into dense ID-embedding tables. However, as embedding tables grow, closed-form solutions become impractical, often necessitating the use of mini-batch gradient descent for training. Despite extensive work on designing loss functions to train CF models, we argue that one core component of these pipelines is heavily overlooked: weight decay. Attaining high-performing models typically requires careful tuning of weight decay, regardless of loss, yet its necessity i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.11318","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/2505.11318/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":"2505.11318","created_at":"2026-07-05T12:02:46.239636+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.11318v2","created_at":"2026-07-05T12:02:46.239636+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.11318","created_at":"2026-07-05T12:02:46.239636+00:00"},{"alias_kind":"pith_short_12","alias_value":"7OMXF6P4JLYN","created_at":"2026-07-05T12:02:46.239636+00:00"},{"alias_kind":"pith_short_16","alias_value":"7OMXF6P4JLYNAVHE","created_at":"2026-07-05T12:02:46.239636+00:00"},{"alias_kind":"pith_short_8","alias_value":"7OMXF6P4","created_at":"2026-07-05T12:02:46.239636+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.22224","citing_title":"Generative Recommendation with Semantic IDs: A Practitioner's Handbook","ref_index":42,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7OMXF6P4JLYNAVHEKVI7UYOGDT","json":"https://pith.science/pith/7OMXF6P4JLYNAVHEKVI7UYOGDT.json","graph_json":"https://pith.science/api/pith-number/7OMXF6P4JLYNAVHEKVI7UYOGDT/graph.json","events_json":"https://pith.science/api/pith-number/7OMXF6P4JLYNAVHEKVI7UYOGDT/events.json","paper":"https://pith.science/paper/7OMXF6P4"},"agent_actions":{"view_html":"https://pith.science/pith/7OMXF6P4JLYNAVHEKVI7UYOGDT","download_json":"https://pith.science/pith/7OMXF6P4JLYNAVHEKVI7UYOGDT.json","view_paper":"https://pith.science/paper/7OMXF6P4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.11318&json=true","fetch_graph":"https://pith.science/api/pith-number/7OMXF6P4JLYNAVHEKVI7UYOGDT/graph.json","fetch_events":"https://pith.science/api/pith-number/7OMXF6P4JLYNAVHEKVI7UYOGDT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7OMXF6P4JLYNAVHEKVI7UYOGDT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7OMXF6P4JLYNAVHEKVI7UYOGDT/action/storage_attestation","attest_author":"https://pith.science/pith/7OMXF6P4JLYNAVHEKVI7UYOGDT/action/author_attestation","sign_citation":"https://pith.science/pith/7OMXF6P4JLYNAVHEKVI7UYOGDT/action/citation_signature","submit_replication":"https://pith.science/pith/7OMXF6P4JLYNAVHEKVI7UYOGDT/action/replication_record"}},"created_at":"2026-07-05T12:02:46.239636+00:00","updated_at":"2026-07-05T12:02:46.239636+00:00"}