{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:2FUFN672PMQZMHTHFBEVFYIKPW","short_pith_number":"pith:2FUFN672","schema_version":"1.0","canonical_sha256":"d16856fbfa7b21961e67284952e10a7d92ed2a541096343cbd26804c904d8908","source":{"kind":"arxiv","id":"1712.01807","version":1},"attestation_state":"computed","paper":{"title":"Improving the Performance of Online Neural Transducer Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["eess.AS","stat.ML"],"primary_cat":"cs.CL","authors_text":"Anjuli Kannan, Chung-Cheng Chiu, Patrick Nguyen, Rohit Prabhavalkar, Tara N. Sainath, Yonghui Wu, Zhifeng Chen","submitted_at":"2017-12-05T18:34:56Z","abstract_excerpt":"Having a sequence-to-sequence model which can operate in an online fashion is important for streaming applications such as Voice Search. Neural transducer is a streaming sequence-to-sequence model, but has shown a significant degradation in performance compared to non-streaming models such as Listen, Attend and Spell (LAS). In this paper, we present various improvements to NT. Specifically, we look at increasing the window over which NT computes attention, mainly by looking backwards in time so the model still remains online. In addition, we explore initializing a NT model from a LAS-trained 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":"1712.01807","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2017-12-05T18:34:56Z","cross_cats_sorted":["eess.AS","stat.ML"],"title_canon_sha256":"b12e2d4c066a0a4cc55470cad8d9679c6f817cbd1f1302326bce800eb9f8efaf","abstract_canon_sha256":"94b8fa717edbcce97c4c0717d8c8010a259716ed34c1defcb525bad2dfa1e41a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:28:47.469369Z","signature_b64":"g4vOUePRvMiS0RYZ5dMdlcCOQVZigZKEFgJimZ28YXCLMSxkUXvOE2KTFQBiXC3lGYlGLBG72R1fS9aXq9GdAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d16856fbfa7b21961e67284952e10a7d92ed2a541096343cbd26804c904d8908","last_reissued_at":"2026-05-18T00:28:47.468698Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:28:47.468698Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving the Performance of Online Neural Transducer Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["eess.AS","stat.ML"],"primary_cat":"cs.CL","authors_text":"Anjuli Kannan, Chung-Cheng Chiu, Patrick Nguyen, Rohit Prabhavalkar, Tara N. Sainath, Yonghui Wu, Zhifeng Chen","submitted_at":"2017-12-05T18:34:56Z","abstract_excerpt":"Having a sequence-to-sequence model which can operate in an online fashion is important for streaming applications such as Voice Search. Neural transducer is a streaming sequence-to-sequence model, but has shown a significant degradation in performance compared to non-streaming models such as Listen, Attend and Spell (LAS). In this paper, we present various improvements to NT. Specifically, we look at increasing the window over which NT computes attention, mainly by looking backwards in time so the model still remains online. In addition, we explore initializing a NT model from a LAS-trained m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1712.01807","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":""},"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":"1712.01807","created_at":"2026-05-18T00:28:47.468788+00:00"},{"alias_kind":"arxiv_version","alias_value":"1712.01807v1","created_at":"2026-05-18T00:28:47.468788+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1712.01807","created_at":"2026-05-18T00:28:47.468788+00:00"},{"alias_kind":"pith_short_12","alias_value":"2FUFN672PMQZ","created_at":"2026-05-18T12:30:55.937587+00:00"},{"alias_kind":"pith_short_16","alias_value":"2FUFN672PMQZMHTH","created_at":"2026-05-18T12:30:55.937587+00:00"},{"alias_kind":"pith_short_8","alias_value":"2FUFN672","created_at":"2026-05-18T12:30:55.937587+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/2FUFN672PMQZMHTHFBEVFYIKPW","json":"https://pith.science/pith/2FUFN672PMQZMHTHFBEVFYIKPW.json","graph_json":"https://pith.science/api/pith-number/2FUFN672PMQZMHTHFBEVFYIKPW/graph.json","events_json":"https://pith.science/api/pith-number/2FUFN672PMQZMHTHFBEVFYIKPW/events.json","paper":"https://pith.science/paper/2FUFN672"},"agent_actions":{"view_html":"https://pith.science/pith/2FUFN672PMQZMHTHFBEVFYIKPW","download_json":"https://pith.science/pith/2FUFN672PMQZMHTHFBEVFYIKPW.json","view_paper":"https://pith.science/paper/2FUFN672","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1712.01807&json=true","fetch_graph":"https://pith.science/api/pith-number/2FUFN672PMQZMHTHFBEVFYIKPW/graph.json","fetch_events":"https://pith.science/api/pith-number/2FUFN672PMQZMHTHFBEVFYIKPW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2FUFN672PMQZMHTHFBEVFYIKPW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2FUFN672PMQZMHTHFBEVFYIKPW/action/storage_attestation","attest_author":"https://pith.science/pith/2FUFN672PMQZMHTHFBEVFYIKPW/action/author_attestation","sign_citation":"https://pith.science/pith/2FUFN672PMQZMHTHFBEVFYIKPW/action/citation_signature","submit_replication":"https://pith.science/pith/2FUFN672PMQZMHTHFBEVFYIKPW/action/replication_record"}},"created_at":"2026-05-18T00:28:47.468788+00:00","updated_at":"2026-05-18T00:28:47.468788+00:00"}