{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:V5P5FHTTCEOO4MCSAKWLTN7SEX","short_pith_number":"pith:V5P5FHTT","schema_version":"1.0","canonical_sha256":"af5fd29e73111cee305202acb9b7f225cb207c2b11be80cac1f4b947643bfa8b","source":{"kind":"arxiv","id":"1802.08375","version":2},"attestation_state":"computed","paper":{"title":"Reusing Weights in Subword-aware Neural Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.NE","stat.ML"],"primary_cat":"cs.CL","authors_text":"Rustem Takhanov, Zhenisbek Assylbekov","submitted_at":"2018-02-23T03:44:15Z","abstract_excerpt":"We propose several ways of reusing subword embeddings and other weights in subword-aware neural language models. The proposed techniques do not benefit a competitive character-aware model, but some of them improve the performance of syllable- and morpheme-aware models while showing significant reductions in model sizes. We discover a simple hands-on principle: in a multi-layer input embedding model, layers should be tied consecutively bottom-up if reused at output. Our best morpheme-aware model with properly reused weights beats the competitive word-level model by a large margin across multipl"},"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":"1802.08375","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2018-02-23T03:44:15Z","cross_cats_sorted":["cs.NE","stat.ML"],"title_canon_sha256":"42e985098d0e047e1e34798a7048bfdc2a94fca5b22e4a5a4f569308da893696","abstract_canon_sha256":"056be2d5fdc8a620ddaedc831e895e9fd9a2f468b2b63f307cadfd7e9ee112ac"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:17:32.749587Z","signature_b64":"xX4ObChfGbdM4NhNJ02AuZCyyE/GHzNHBBtx2+16m9Rsk4drUVVevVy9I7W+zEAWEtjVIAEHmLcUgInzwGlCBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"af5fd29e73111cee305202acb9b7f225cb207c2b11be80cac1f4b947643bfa8b","last_reissued_at":"2026-05-18T00:17:32.748937Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:17:32.748937Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reusing Weights in Subword-aware Neural Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.NE","stat.ML"],"primary_cat":"cs.CL","authors_text":"Rustem Takhanov, Zhenisbek Assylbekov","submitted_at":"2018-02-23T03:44:15Z","abstract_excerpt":"We propose several ways of reusing subword embeddings and other weights in subword-aware neural language models. The proposed techniques do not benefit a competitive character-aware model, but some of them improve the performance of syllable- and morpheme-aware models while showing significant reductions in model sizes. We discover a simple hands-on principle: in a multi-layer input embedding model, layers should be tied consecutively bottom-up if reused at output. Our best morpheme-aware model with properly reused weights beats the competitive word-level model by a large margin across multipl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1802.08375","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":""},"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":"1802.08375","created_at":"2026-05-18T00:17:32.749031+00:00"},{"alias_kind":"arxiv_version","alias_value":"1802.08375v2","created_at":"2026-05-18T00:17:32.749031+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1802.08375","created_at":"2026-05-18T00:17:32.749031+00:00"},{"alias_kind":"pith_short_12","alias_value":"V5P5FHTTCEOO","created_at":"2026-05-18T12:32:59.047623+00:00"},{"alias_kind":"pith_short_16","alias_value":"V5P5FHTTCEOO4MCS","created_at":"2026-05-18T12:32:59.047623+00:00"},{"alias_kind":"pith_short_8","alias_value":"V5P5FHTT","created_at":"2026-05-18T12:32:59.047623+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/V5P5FHTTCEOO4MCSAKWLTN7SEX","json":"https://pith.science/pith/V5P5FHTTCEOO4MCSAKWLTN7SEX.json","graph_json":"https://pith.science/api/pith-number/V5P5FHTTCEOO4MCSAKWLTN7SEX/graph.json","events_json":"https://pith.science/api/pith-number/V5P5FHTTCEOO4MCSAKWLTN7SEX/events.json","paper":"https://pith.science/paper/V5P5FHTT"},"agent_actions":{"view_html":"https://pith.science/pith/V5P5FHTTCEOO4MCSAKWLTN7SEX","download_json":"https://pith.science/pith/V5P5FHTTCEOO4MCSAKWLTN7SEX.json","view_paper":"https://pith.science/paper/V5P5FHTT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1802.08375&json=true","fetch_graph":"https://pith.science/api/pith-number/V5P5FHTTCEOO4MCSAKWLTN7SEX/graph.json","fetch_events":"https://pith.science/api/pith-number/V5P5FHTTCEOO4MCSAKWLTN7SEX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/V5P5FHTTCEOO4MCSAKWLTN7SEX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/V5P5FHTTCEOO4MCSAKWLTN7SEX/action/storage_attestation","attest_author":"https://pith.science/pith/V5P5FHTTCEOO4MCSAKWLTN7SEX/action/author_attestation","sign_citation":"https://pith.science/pith/V5P5FHTTCEOO4MCSAKWLTN7SEX/action/citation_signature","submit_replication":"https://pith.science/pith/V5P5FHTTCEOO4MCSAKWLTN7SEX/action/replication_record"}},"created_at":"2026-05-18T00:17:32.749031+00:00","updated_at":"2026-05-18T00:17:32.749031+00:00"}