{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:UYOUYI6IUCTWAHROFJDFRB4PKD","short_pith_number":"pith:UYOUYI6I","schema_version":"1.0","canonical_sha256":"a61d4c23c8a0a7601e2e2a4658878f50da258b1ec03d02174ce6c323fe54acc3","source":{"kind":"arxiv","id":"1910.06720","version":2},"attestation_state":"computed","paper":{"title":"Improving Word Embedding Factorization for Compression Using Distilled Nonlinear Neural Decomposition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Ahmad Rashid, Krtin Kumar, Md Akmal Haidar, Mehdi Rezagholizadeh, Vasileios Lioutas","submitted_at":"2019-10-02T16:40:03Z","abstract_excerpt":"Word-embeddings are vital components of Natural Language Processing (NLP) models and have been extensively explored. However, they consume a lot of memory which poses a challenge for edge deployment. Embedding matrices, typically, contain most of the parameters for language models and about a third for machine translation systems. In this paper, we propose Distilled Embedding, an (input/output) embedding compression method based on low-rank matrix decomposition and knowledge distillation. First, we initialize the weights of our decomposed matrices by learning to reconstruct the full pre-traine"},"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":"1910.06720","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-10-02T16:40:03Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"be44319e738c997c7a9acf98a3941f5af7777a605d9889d7aad4cf4eba44b8b2","abstract_canon_sha256":"e2196595a6ac8a28f228432518b3ece247d69507411c8b3093014709f7afefca"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:50:46.460550Z","signature_b64":"ox0eogWEngEiFHnllqaHB5gXxJcvzi37cQoOJuiHOGNPCwxu74zLQP7yEXGRwipUSVTtRawWaQofr5DXvlLKAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a61d4c23c8a0a7601e2e2a4658878f50da258b1ec03d02174ce6c323fe54acc3","last_reissued_at":"2026-07-05T01:50:46.460066Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:50:46.460066Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Word Embedding Factorization for Compression Using Distilled Nonlinear Neural Decomposition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Ahmad Rashid, Krtin Kumar, Md Akmal Haidar, Mehdi Rezagholizadeh, Vasileios Lioutas","submitted_at":"2019-10-02T16:40:03Z","abstract_excerpt":"Word-embeddings are vital components of Natural Language Processing (NLP) models and have been extensively explored. However, they consume a lot of memory which poses a challenge for edge deployment. Embedding matrices, typically, contain most of the parameters for language models and about a third for machine translation systems. In this paper, we propose Distilled Embedding, an (input/output) embedding compression method based on low-rank matrix decomposition and knowledge distillation. First, we initialize the weights of our decomposed matrices by learning to reconstruct the full pre-traine"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1910.06720","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/1910.06720/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":"1910.06720","created_at":"2026-07-05T01:50:46.460123+00:00"},{"alias_kind":"arxiv_version","alias_value":"1910.06720v2","created_at":"2026-07-05T01:50:46.460123+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1910.06720","created_at":"2026-07-05T01:50:46.460123+00:00"},{"alias_kind":"pith_short_12","alias_value":"UYOUYI6IUCTW","created_at":"2026-07-05T01:50:46.460123+00:00"},{"alias_kind":"pith_short_16","alias_value":"UYOUYI6IUCTWAHRO","created_at":"2026-07-05T01:50:46.460123+00:00"},{"alias_kind":"pith_short_8","alias_value":"UYOUYI6I","created_at":"2026-07-05T01:50:46.460123+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.10453","citing_title":"SlimSpec: Low-Rank Draft LM-Head for Accelerated Speculative Decoding","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UYOUYI6IUCTWAHROFJDFRB4PKD","json":"https://pith.science/pith/UYOUYI6IUCTWAHROFJDFRB4PKD.json","graph_json":"https://pith.science/api/pith-number/UYOUYI6IUCTWAHROFJDFRB4PKD/graph.json","events_json":"https://pith.science/api/pith-number/UYOUYI6IUCTWAHROFJDFRB4PKD/events.json","paper":"https://pith.science/paper/UYOUYI6I"},"agent_actions":{"view_html":"https://pith.science/pith/UYOUYI6IUCTWAHROFJDFRB4PKD","download_json":"https://pith.science/pith/UYOUYI6IUCTWAHROFJDFRB4PKD.json","view_paper":"https://pith.science/paper/UYOUYI6I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1910.06720&json=true","fetch_graph":"https://pith.science/api/pith-number/UYOUYI6IUCTWAHROFJDFRB4PKD/graph.json","fetch_events":"https://pith.science/api/pith-number/UYOUYI6IUCTWAHROFJDFRB4PKD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UYOUYI6IUCTWAHROFJDFRB4PKD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UYOUYI6IUCTWAHROFJDFRB4PKD/action/storage_attestation","attest_author":"https://pith.science/pith/UYOUYI6IUCTWAHROFJDFRB4PKD/action/author_attestation","sign_citation":"https://pith.science/pith/UYOUYI6IUCTWAHROFJDFRB4PKD/action/citation_signature","submit_replication":"https://pith.science/pith/UYOUYI6IUCTWAHROFJDFRB4PKD/action/replication_record"}},"created_at":"2026-07-05T01:50:46.460123+00:00","updated_at":"2026-07-05T01:50:46.460123+00:00"}