{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:3C7ESQ56CFWSNQAZAYDQLGZVM7","short_pith_number":"pith:3C7ESQ56","schema_version":"1.0","canonical_sha256":"d8be4943be116d26c0190607059b3567db81bbcf8f913c9d68a63cf50d78646d","source":{"kind":"arxiv","id":"2006.05623","version":1},"attestation_state":"computed","paper":{"title":"Training with Multi-Layer Embeddings for Model Reduction","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Ashish Kumar Singh, Benjamin Cho, Benjamin Ghaemmaghami, Leo Orshansky, Mattan Erez, Michael Orshansky, Zihao Deng","submitted_at":"2020-06-10T02:47:40Z","abstract_excerpt":"Modern recommendation systems rely on real-valued embeddings of categorical features. Increasing the dimension of embedding vectors improves model accuracy but comes at a high cost to model size. We introduce a multi-layer embedding training (MLET) architecture that trains embeddings via a sequence of linear layers to derive superior embedding accuracy vs. model size trade-off.\n  Our approach is fundamentally based on the ability of factorized linear layers to produce superior embeddings to that of a single linear layer. We focus on the analysis and implementation of a two-layer scheme. Harnes"},"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":"2006.05623","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-06-10T02:47:40Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"c51531ad52ce0e61283566bb64716e7fe562559003824729e91abe24fdc10ae8","abstract_canon_sha256":"b60eea10cfe0b15e7706584c2719be5134e964e5eb1a504bdea54a8e0ccee693"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:09:15.385733Z","signature_b64":"v47Y2YdkPgs6H99u/hb1IYvx/7B3pmgxS6kwJnDkaJUghIkkfLwW01gP5KWxH+nybBdaOskcGDQaC1E/UCK8Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d8be4943be116d26c0190607059b3567db81bbcf8f913c9d68a63cf50d78646d","last_reissued_at":"2026-07-05T01:09:15.385276Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:09:15.385276Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Training with Multi-Layer Embeddings for Model Reduction","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Ashish Kumar Singh, Benjamin Cho, Benjamin Ghaemmaghami, Leo Orshansky, Mattan Erez, Michael Orshansky, Zihao Deng","submitted_at":"2020-06-10T02:47:40Z","abstract_excerpt":"Modern recommendation systems rely on real-valued embeddings of categorical features. Increasing the dimension of embedding vectors improves model accuracy but comes at a high cost to model size. We introduce a multi-layer embedding training (MLET) architecture that trains embeddings via a sequence of linear layers to derive superior embedding accuracy vs. model size trade-off.\n  Our approach is fundamentally based on the ability of factorized linear layers to produce superior embeddings to that of a single linear layer. We focus on the analysis and implementation of a two-layer scheme. Harnes"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.05623","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/2006.05623/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":"2006.05623","created_at":"2026-07-05T01:09:15.385340+00:00"},{"alias_kind":"arxiv_version","alias_value":"2006.05623v1","created_at":"2026-07-05T01:09:15.385340+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.05623","created_at":"2026-07-05T01:09:15.385340+00:00"},{"alias_kind":"pith_short_12","alias_value":"3C7ESQ56CFWS","created_at":"2026-07-05T01:09:15.385340+00:00"},{"alias_kind":"pith_short_16","alias_value":"3C7ESQ56CFWSNQAZ","created_at":"2026-07-05T01:09:15.385340+00:00"},{"alias_kind":"pith_short_8","alias_value":"3C7ESQ56","created_at":"2026-07-05T01:09:15.385340+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/3C7ESQ56CFWSNQAZAYDQLGZVM7","json":"https://pith.science/pith/3C7ESQ56CFWSNQAZAYDQLGZVM7.json","graph_json":"https://pith.science/api/pith-number/3C7ESQ56CFWSNQAZAYDQLGZVM7/graph.json","events_json":"https://pith.science/api/pith-number/3C7ESQ56CFWSNQAZAYDQLGZVM7/events.json","paper":"https://pith.science/paper/3C7ESQ56"},"agent_actions":{"view_html":"https://pith.science/pith/3C7ESQ56CFWSNQAZAYDQLGZVM7","download_json":"https://pith.science/pith/3C7ESQ56CFWSNQAZAYDQLGZVM7.json","view_paper":"https://pith.science/paper/3C7ESQ56","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2006.05623&json=true","fetch_graph":"https://pith.science/api/pith-number/3C7ESQ56CFWSNQAZAYDQLGZVM7/graph.json","fetch_events":"https://pith.science/api/pith-number/3C7ESQ56CFWSNQAZAYDQLGZVM7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3C7ESQ56CFWSNQAZAYDQLGZVM7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3C7ESQ56CFWSNQAZAYDQLGZVM7/action/storage_attestation","attest_author":"https://pith.science/pith/3C7ESQ56CFWSNQAZAYDQLGZVM7/action/author_attestation","sign_citation":"https://pith.science/pith/3C7ESQ56CFWSNQAZAYDQLGZVM7/action/citation_signature","submit_replication":"https://pith.science/pith/3C7ESQ56CFWSNQAZAYDQLGZVM7/action/replication_record"}},"created_at":"2026-07-05T01:09:15.385340+00:00","updated_at":"2026-07-05T01:09:15.385340+00:00"}