{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:5XAWQHPFWGJ6QAWSHEJDXNSUYZ","short_pith_number":"pith:5XAWQHPF","schema_version":"1.0","canonical_sha256":"edc1681de5b193e802d239123bb654c64195adca6dc81beb836cff5415a7644c","source":{"kind":"arxiv","id":"2310.04400","version":2},"attestation_state":"computed","paper":{"title":"On the Embedding Collapse when Scaling up Recommendation Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.LG","authors_text":"Baixu Chen, Jie Jiang, Junwei Pan, Mingsheng Long, Ximei Wang, Xingzhuo Guo","submitted_at":"2023-10-06T17:50:38Z","abstract_excerpt":"Recent advances in foundation models have led to a promising trend of developing large recommendation models to leverage vast amounts of available data. Still, mainstream models remain embarrassingly small in size and na\\\"ive enlarging does not lead to sufficient performance gain, suggesting a deficiency in the model scalability. In this paper, we identify the embedding collapse phenomenon as the inhibition of scalability, wherein the embedding matrix tends to occupy a low-dimensional subspace. Through empirical and theoretical analysis, we demonstrate a \\emph{two-sided effect} of feature inte"},"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":"2310.04400","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-10-06T17:50:38Z","cross_cats_sorted":["cs.IR"],"title_canon_sha256":"06cd2df9640f106d7d0b9e6494efaa9bb605f79f391ed56785aab37bfa18a8b4","abstract_canon_sha256":"4d29d5d9dc9754cabef2595fee637682966f1ffcbfa91cd2241b518814794078"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:28:02.860031Z","signature_b64":"AsLO0EaLb4uXBhhFweLru0/6mOledeZqSn6fDtLO8ZgkmkiobYTffWiLrCIivhXNBvoy8hB9MMCVRceOOh53AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"edc1681de5b193e802d239123bb654c64195adca6dc81beb836cff5415a7644c","last_reissued_at":"2026-07-05T08:28:02.859544Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:28:02.859544Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On the Embedding Collapse when Scaling up Recommendation Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.LG","authors_text":"Baixu Chen, Jie Jiang, Junwei Pan, Mingsheng Long, Ximei Wang, Xingzhuo Guo","submitted_at":"2023-10-06T17:50:38Z","abstract_excerpt":"Recent advances in foundation models have led to a promising trend of developing large recommendation models to leverage vast amounts of available data. Still, mainstream models remain embarrassingly small in size and na\\\"ive enlarging does not lead to sufficient performance gain, suggesting a deficiency in the model scalability. In this paper, we identify the embedding collapse phenomenon as the inhibition of scalability, wherein the embedding matrix tends to occupy a low-dimensional subspace. Through empirical and theoretical analysis, we demonstrate a \\emph{two-sided effect} of feature inte"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.04400","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/2310.04400/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":"2310.04400","created_at":"2026-07-05T08:28:02.859602+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.04400v2","created_at":"2026-07-05T08:28:02.859602+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.04400","created_at":"2026-07-05T08:28:02.859602+00:00"},{"alias_kind":"pith_short_12","alias_value":"5XAWQHPFWGJ6","created_at":"2026-07-05T08:28:02.859602+00:00"},{"alias_kind":"pith_short_16","alias_value":"5XAWQHPFWGJ6QAWS","created_at":"2026-07-05T08:28:02.859602+00:00"},{"alias_kind":"pith_short_8","alias_value":"5XAWQHPF","created_at":"2026-07-05T08:28:02.859602+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21911","citing_title":"The Pitfall of Scaling Up: Uncovering and Mitigating Popularity Bias Amplification in Scaling Transformer-based Recommenders","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07599","citing_title":"DiffoR: A Unified Continuous Generative Framework for Universal Ordinal Regression","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2602.17050","citing_title":"Multi-Probe Zero Collision Hash (MPZCH): Mitigating Embedding Collisions and Enhancing Model Freshness in Large-Scale Recommenders","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10886","citing_title":"LoKA: Low-precision Kernel Applications for Recommendation Models At Scale","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17878","citing_title":"RankUp: Towards High-rank Representations for Large Scale Advertising Recommender Systems","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10886","citing_title":"LoKA: Low-precision Kernel Applications for Recommendation Models At Scale","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17878","citing_title":"RankUp: Towards High-rank Representations for Large Scale Advertising Recommender Systems","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5XAWQHPFWGJ6QAWSHEJDXNSUYZ","json":"https://pith.science/pith/5XAWQHPFWGJ6QAWSHEJDXNSUYZ.json","graph_json":"https://pith.science/api/pith-number/5XAWQHPFWGJ6QAWSHEJDXNSUYZ/graph.json","events_json":"https://pith.science/api/pith-number/5XAWQHPFWGJ6QAWSHEJDXNSUYZ/events.json","paper":"https://pith.science/paper/5XAWQHPF"},"agent_actions":{"view_html":"https://pith.science/pith/5XAWQHPFWGJ6QAWSHEJDXNSUYZ","download_json":"https://pith.science/pith/5XAWQHPFWGJ6QAWSHEJDXNSUYZ.json","view_paper":"https://pith.science/paper/5XAWQHPF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.04400&json=true","fetch_graph":"https://pith.science/api/pith-number/5XAWQHPFWGJ6QAWSHEJDXNSUYZ/graph.json","fetch_events":"https://pith.science/api/pith-number/5XAWQHPFWGJ6QAWSHEJDXNSUYZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5XAWQHPFWGJ6QAWSHEJDXNSUYZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5XAWQHPFWGJ6QAWSHEJDXNSUYZ/action/storage_attestation","attest_author":"https://pith.science/pith/5XAWQHPFWGJ6QAWSHEJDXNSUYZ/action/author_attestation","sign_citation":"https://pith.science/pith/5XAWQHPFWGJ6QAWSHEJDXNSUYZ/action/citation_signature","submit_replication":"https://pith.science/pith/5XAWQHPFWGJ6QAWSHEJDXNSUYZ/action/replication_record"}},"created_at":"2026-07-05T08:28:02.859602+00:00","updated_at":"2026-07-05T08:28:02.859602+00:00"}