{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:MI6KQB44HPXRXB4FCKGR7LEFKV","short_pith_number":"pith:MI6KQB44","schema_version":"1.0","canonical_sha256":"623ca8079c3bef1b8785128d1fac85556d9f41652d5440fb1e0933b338305224","source":{"kind":"arxiv","id":"2006.05165","version":1},"attestation_state":"computed","paper":{"title":"Learning to Recover from Multi-Modality Errors for Non-Autoregressive Neural Machine Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jie Zhou, Peng Li, Qiu Ran, Yankai Lin","submitted_at":"2020-06-09T10:12:16Z","abstract_excerpt":"Non-autoregressive neural machine translation (NAT) predicts the entire target sequence simultaneously and significantly accelerates inference process. However, NAT discards the dependency information in a sentence, and thus inevitably suffers from the multi-modality problem: the target tokens may be provided by different possible translations, often causing token repetitions or missing. To alleviate this problem, we propose a novel semi-autoregressive model RecoverSAT in this work, which generates a translation as a sequence of segments. The segments are generated simultaneously while each se"},"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":"2006.05165","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-06-09T10:12:16Z","cross_cats_sorted":[],"title_canon_sha256":"6fcf3954146b20ceb39ec349fc4ff196a4c6bd55d41a476b67e4132b9ba2871a","abstract_canon_sha256":"726ec218e059bd7acb5232e739dfc3d983fa7755eef7c8e29e48341ce7428779"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:09:11.232484Z","signature_b64":"g+KGiLUH4355NGYB3P0Bbic3Rf0AgPINmxF5fe/RQ5d8ykEzDtBeLHgxHmcmJEyOyjG3QU67Fx+WxwtEfiu3DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"623ca8079c3bef1b8785128d1fac85556d9f41652d5440fb1e0933b338305224","last_reissued_at":"2026-07-05T01:09:11.232002Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:09:11.232002Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning to Recover from Multi-Modality Errors for Non-Autoregressive Neural Machine Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jie Zhou, Peng Li, Qiu Ran, Yankai Lin","submitted_at":"2020-06-09T10:12:16Z","abstract_excerpt":"Non-autoregressive neural machine translation (NAT) predicts the entire target sequence simultaneously and significantly accelerates inference process. However, NAT discards the dependency information in a sentence, and thus inevitably suffers from the multi-modality problem: the target tokens may be provided by different possible translations, often causing token repetitions or missing. To alleviate this problem, we propose a novel semi-autoregressive model RecoverSAT in this work, which generates a translation as a sequence of segments. The segments are generated simultaneously while each se"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.05165","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2006.05165/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":"2006.05165","created_at":"2026-07-05T01:09:11.232064+00:00"},{"alias_kind":"arxiv_version","alias_value":"2006.05165v1","created_at":"2026-07-05T01:09:11.232064+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.05165","created_at":"2026-07-05T01:09:11.232064+00:00"},{"alias_kind":"pith_short_12","alias_value":"MI6KQB44HPXR","created_at":"2026-07-05T01:09:11.232064+00:00"},{"alias_kind":"pith_short_16","alias_value":"MI6KQB44HPXRXB4F","created_at":"2026-07-05T01:09:11.232064+00:00"},{"alias_kind":"pith_short_8","alias_value":"MI6KQB44","created_at":"2026-07-05T01:09:11.232064+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.15152","citing_title":"Sculpting Features from Noise: Reward-Guided Hierarchical Diffusion for Task-Optimal Feature Transformation","ref_index":22,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MI6KQB44HPXRXB4FCKGR7LEFKV","json":"https://pith.science/pith/MI6KQB44HPXRXB4FCKGR7LEFKV.json","graph_json":"https://pith.science/api/pith-number/MI6KQB44HPXRXB4FCKGR7LEFKV/graph.json","events_json":"https://pith.science/api/pith-number/MI6KQB44HPXRXB4FCKGR7LEFKV/events.json","paper":"https://pith.science/paper/MI6KQB44"},"agent_actions":{"view_html":"https://pith.science/pith/MI6KQB44HPXRXB4FCKGR7LEFKV","download_json":"https://pith.science/pith/MI6KQB44HPXRXB4FCKGR7LEFKV.json","view_paper":"https://pith.science/paper/MI6KQB44","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2006.05165&json=true","fetch_graph":"https://pith.science/api/pith-number/MI6KQB44HPXRXB4FCKGR7LEFKV/graph.json","fetch_events":"https://pith.science/api/pith-number/MI6KQB44HPXRXB4FCKGR7LEFKV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MI6KQB44HPXRXB4FCKGR7LEFKV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MI6KQB44HPXRXB4FCKGR7LEFKV/action/storage_attestation","attest_author":"https://pith.science/pith/MI6KQB44HPXRXB4FCKGR7LEFKV/action/author_attestation","sign_citation":"https://pith.science/pith/MI6KQB44HPXRXB4FCKGR7LEFKV/action/citation_signature","submit_replication":"https://pith.science/pith/MI6KQB44HPXRXB4FCKGR7LEFKV/action/replication_record"}},"created_at":"2026-07-05T01:09:11.232064+00:00","updated_at":"2026-07-05T01:09:11.232064+00:00"}