{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:4T3A6K4ODPQBYN5IPX27LEYJ62","short_pith_number":"pith:4T3A6K4O","schema_version":"1.0","canonical_sha256":"e4f60f2b8e1be01c37a87df5f59309f6a52f20b8765db31f44e8a9e97a7f0e96","source":{"kind":"arxiv","id":"1908.10118","version":2},"attestation_state":"computed","paper":{"title":"Multi-Layer Softmaxing during Training Neural Machine Translation for Flexible Decoding with Fewer Layers","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Atsushi Fujita, Raj Dabre","submitted_at":"2019-08-27T10:17:24Z","abstract_excerpt":"This paper proposes a novel procedure for training an encoder-decoder based deep neural network which compresses NxM models into a single model enabling us to dynamically choose the number of encoder and decoder layers for decoding. Usually, the output of the last layer of the N-layer encoder is fed to the M-layer decoder, and the output of the last decoder layer is used to compute softmax loss. Instead, our method computes a single loss consisting of NxM losses: the softmax loss for the output of each of the M decoder layers derived using the output of each of the N encoder layers. A single m"},"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":"1908.10118","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2019-08-27T10:17:24Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"b22c0fe983295ee26bfe556f84d27f0cf20ff561d237d12b1336cba9ef017933","abstract_canon_sha256":"9607a64f427a517ded2d61152348a229eeca4f00712c4b9d0608bd7d1f483e1d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:00:16.518749Z","signature_b64":"tF1o640w1Bc+7iVGw2Ny/vETqv/V1+EJ73io9WWqN7XchxJ7G3TVX9tYsEeQ3WFiuCsiDtUSN8h09cu0wo9lAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e4f60f2b8e1be01c37a87df5f59309f6a52f20b8765db31f44e8a9e97a7f0e96","last_reissued_at":"2026-07-05T00:00:16.518359Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:00:16.518359Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multi-Layer Softmaxing during Training Neural Machine Translation for Flexible Decoding with Fewer Layers","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Atsushi Fujita, Raj Dabre","submitted_at":"2019-08-27T10:17:24Z","abstract_excerpt":"This paper proposes a novel procedure for training an encoder-decoder based deep neural network which compresses NxM models into a single model enabling us to dynamically choose the number of encoder and decoder layers for decoding. Usually, the output of the last layer of the N-layer encoder is fed to the M-layer decoder, and the output of the last decoder layer is used to compute softmax loss. Instead, our method computes a single loss consisting of NxM losses: the softmax loss for the output of each of the M decoder layers derived using the output of each of the N encoder layers. A single m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1908.10118","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/1908.10118/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":"1908.10118","created_at":"2026-07-05T00:00:16.518417+00:00"},{"alias_kind":"arxiv_version","alias_value":"1908.10118v2","created_at":"2026-07-05T00:00:16.518417+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1908.10118","created_at":"2026-07-05T00:00:16.518417+00:00"},{"alias_kind":"pith_short_12","alias_value":"4T3A6K4ODPQB","created_at":"2026-07-05T00:00:16.518417+00:00"},{"alias_kind":"pith_short_16","alias_value":"4T3A6K4ODPQBYN5I","created_at":"2026-07-05T00:00:16.518417+00:00"},{"alias_kind":"pith_short_8","alias_value":"4T3A6K4O","created_at":"2026-07-05T00:00:16.518417+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/4T3A6K4ODPQBYN5IPX27LEYJ62","json":"https://pith.science/pith/4T3A6K4ODPQBYN5IPX27LEYJ62.json","graph_json":"https://pith.science/api/pith-number/4T3A6K4ODPQBYN5IPX27LEYJ62/graph.json","events_json":"https://pith.science/api/pith-number/4T3A6K4ODPQBYN5IPX27LEYJ62/events.json","paper":"https://pith.science/paper/4T3A6K4O"},"agent_actions":{"view_html":"https://pith.science/pith/4T3A6K4ODPQBYN5IPX27LEYJ62","download_json":"https://pith.science/pith/4T3A6K4ODPQBYN5IPX27LEYJ62.json","view_paper":"https://pith.science/paper/4T3A6K4O","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1908.10118&json=true","fetch_graph":"https://pith.science/api/pith-number/4T3A6K4ODPQBYN5IPX27LEYJ62/graph.json","fetch_events":"https://pith.science/api/pith-number/4T3A6K4ODPQBYN5IPX27LEYJ62/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4T3A6K4ODPQBYN5IPX27LEYJ62/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4T3A6K4ODPQBYN5IPX27LEYJ62/action/storage_attestation","attest_author":"https://pith.science/pith/4T3A6K4ODPQBYN5IPX27LEYJ62/action/author_attestation","sign_citation":"https://pith.science/pith/4T3A6K4ODPQBYN5IPX27LEYJ62/action/citation_signature","submit_replication":"https://pith.science/pith/4T3A6K4ODPQBYN5IPX27LEYJ62/action/replication_record"}},"created_at":"2026-07-05T00:00:16.518417+00:00","updated_at":"2026-07-05T00:00:16.518417+00:00"}