{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:OUS2V7WTSEBXE5QB5DD5D5UP5V","short_pith_number":"pith:OUS2V7WT","schema_version":"1.0","canonical_sha256":"7525aafed39103727601e8c7d1f68fed7cabb0f3108c2ff24784322611ab78ab","source":{"kind":"arxiv","id":"1805.05225","version":2},"attestation_state":"computed","paper":{"title":"RETURNN as a Generic Flexible Neural Toolkit with Application to Translation and Speech Recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.NE","authors_text":"Albert Zeyer, Hermann Ney, Tamer Alkhouli","submitted_at":"2018-05-14T15:23:40Z","abstract_excerpt":"We compare the fast training and decoding speed of RETURNN of attention models for translation, due to fast CUDA LSTM kernels, and a fast pure TensorFlow beam search decoder. We show that a layer-wise pretraining scheme for recurrent attention models gives over 1% BLEU improvement absolute and it allows to train deeper recurrent encoder networks. Promising preliminary results on max. expected BLEU training are presented. We are able to train state-of-the-art models for translation and end-to-end models for speech recognition and show results on WMT 2017 and Switchboard. The flexibility of RETU"},"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":"1805.05225","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2018-05-14T15:23:40Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"581b10b0378e5410103e30d29e2f0dc3343b24886b8d829b2d239c3031e02eea","abstract_canon_sha256":"7f6828ce44c25325c6b03c2ddae05a958c176bdef41406429f96ce8b335e89c0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-04T23:51:18.342580Z","signature_b64":"5wInWEodZPFf8fAaWAOrLc9SJ8cI9ChNACM5t483S9UoPiJZvhxAmBHByQG5F5gnZJFOAEw+r+iIdrERDyf3Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7525aafed39103727601e8c7d1f68fed7cabb0f3108c2ff24784322611ab78ab","last_reissued_at":"2026-07-04T23:51:18.342102Z","signature_status":"signed_v1","first_computed_at":"2026-07-04T23:51:18.342102Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RETURNN as a Generic Flexible Neural Toolkit with Application to Translation and Speech Recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.NE","authors_text":"Albert Zeyer, Hermann Ney, Tamer Alkhouli","submitted_at":"2018-05-14T15:23:40Z","abstract_excerpt":"We compare the fast training and decoding speed of RETURNN of attention models for translation, due to fast CUDA LSTM kernels, and a fast pure TensorFlow beam search decoder. We show that a layer-wise pretraining scheme for recurrent attention models gives over 1% BLEU improvement absolute and it allows to train deeper recurrent encoder networks. Promising preliminary results on max. expected BLEU training are presented. We are able to train state-of-the-art models for translation and end-to-end models for speech recognition and show results on WMT 2017 and Switchboard. The flexibility of RETU"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1805.05225","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/1805.05225/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":"1805.05225","created_at":"2026-07-04T23:51:18.342160+00:00"},{"alias_kind":"arxiv_version","alias_value":"1805.05225v2","created_at":"2026-07-04T23:51:18.342160+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1805.05225","created_at":"2026-07-04T23:51:18.342160+00:00"},{"alias_kind":"pith_short_12","alias_value":"OUS2V7WTSEBX","created_at":"2026-07-04T23:51:18.342160+00:00"},{"alias_kind":"pith_short_16","alias_value":"OUS2V7WTSEBXE5QB","created_at":"2026-07-04T23:51:18.342160+00:00"},{"alias_kind":"pith_short_8","alias_value":"OUS2V7WT","created_at":"2026-07-04T23:51:18.342160+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/OUS2V7WTSEBXE5QB5DD5D5UP5V","json":"https://pith.science/pith/OUS2V7WTSEBXE5QB5DD5D5UP5V.json","graph_json":"https://pith.science/api/pith-number/OUS2V7WTSEBXE5QB5DD5D5UP5V/graph.json","events_json":"https://pith.science/api/pith-number/OUS2V7WTSEBXE5QB5DD5D5UP5V/events.json","paper":"https://pith.science/paper/OUS2V7WT"},"agent_actions":{"view_html":"https://pith.science/pith/OUS2V7WTSEBXE5QB5DD5D5UP5V","download_json":"https://pith.science/pith/OUS2V7WTSEBXE5QB5DD5D5UP5V.json","view_paper":"https://pith.science/paper/OUS2V7WT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1805.05225&json=true","fetch_graph":"https://pith.science/api/pith-number/OUS2V7WTSEBXE5QB5DD5D5UP5V/graph.json","fetch_events":"https://pith.science/api/pith-number/OUS2V7WTSEBXE5QB5DD5D5UP5V/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OUS2V7WTSEBXE5QB5DD5D5UP5V/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OUS2V7WTSEBXE5QB5DD5D5UP5V/action/storage_attestation","attest_author":"https://pith.science/pith/OUS2V7WTSEBXE5QB5DD5D5UP5V/action/author_attestation","sign_citation":"https://pith.science/pith/OUS2V7WTSEBXE5QB5DD5D5UP5V/action/citation_signature","submit_replication":"https://pith.science/pith/OUS2V7WTSEBXE5QB5DD5D5UP5V/action/replication_record"}},"created_at":"2026-07-04T23:51:18.342160+00:00","updated_at":"2026-07-04T23:51:18.342160+00:00"}