{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:VHHKA3XC6YF3JDBVZ7NI2JJ62U","short_pith_number":"pith:VHHKA3XC","schema_version":"1.0","canonical_sha256":"a9cea06ee2f60bb48c35cfda8d253ed53171409390e1955aa4b6a51b26a7110e","source":{"kind":"arxiv","id":"2005.00850","version":2},"attestation_state":"computed","paper":{"title":"ENGINE: Energy-Based Inference Networks for Non-Autoregressive Machine Translation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Kevin Gimpel, Lifu Tu, Richard Yuanzhe Pang, Sam Wiseman","submitted_at":"2020-05-02T15:06:47Z","abstract_excerpt":"We propose to train a non-autoregressive machine translation model to minimize the energy defined by a pretrained autoregressive model. In particular, we view our non-autoregressive translation system as an inference network (Tu and Gimpel, 2018) trained to minimize the autoregressive teacher energy. This contrasts with the popular approach of training a non-autoregressive model on a distilled corpus consisting of the beam-searched outputs of such a teacher model. Our approach, which we call ENGINE (ENerGy-based Inference NEtworks), achieves state-of-the-art non-autoregressive results on the I"},"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":"2005.00850","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2020-05-02T15:06:47Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"3b489429b4d5be69efcd1f3efb5dff2b02657c7c59488949ff150f9854c38b8a","abstract_canon_sha256":"55b07c911359496a81a8dda402c8a2df72d3b78cf05bdfc6c376852e77694dab"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:02:30.146062Z","signature_b64":"6cJFU187OaPm/6j3BhMb635tgU3pUvNYwQVJSrkfJZMntn26iFW5BYzy0kftOWgvhoOUCU9yUS/hbeXaVAEqBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a9cea06ee2f60bb48c35cfda8d253ed53171409390e1955aa4b6a51b26a7110e","last_reissued_at":"2026-07-05T01:02:30.145592Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:02:30.145592Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ENGINE: Energy-Based Inference Networks for Non-Autoregressive Machine Translation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Kevin Gimpel, Lifu Tu, Richard Yuanzhe Pang, Sam Wiseman","submitted_at":"2020-05-02T15:06:47Z","abstract_excerpt":"We propose to train a non-autoregressive machine translation model to minimize the energy defined by a pretrained autoregressive model. In particular, we view our non-autoregressive translation system as an inference network (Tu and Gimpel, 2018) trained to minimize the autoregressive teacher energy. This contrasts with the popular approach of training a non-autoregressive model on a distilled corpus consisting of the beam-searched outputs of such a teacher model. Our approach, which we call ENGINE (ENerGy-based Inference NEtworks), achieves state-of-the-art non-autoregressive results on the I"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2005.00850","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/2005.00850/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":"2005.00850","created_at":"2026-07-05T01:02:30.145652+00:00"},{"alias_kind":"arxiv_version","alias_value":"2005.00850v2","created_at":"2026-07-05T01:02:30.145652+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2005.00850","created_at":"2026-07-05T01:02:30.145652+00:00"},{"alias_kind":"pith_short_12","alias_value":"VHHKA3XC6YF3","created_at":"2026-07-05T01:02:30.145652+00:00"},{"alias_kind":"pith_short_16","alias_value":"VHHKA3XC6YF3JDBV","created_at":"2026-07-05T01:02:30.145652+00:00"},{"alias_kind":"pith_short_8","alias_value":"VHHKA3XC","created_at":"2026-07-05T01:02:30.145652+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.14528","citing_title":"Multi-Level Optimal Transport for Universal Cross-Tokenizer Knowledge Distillation on Language Models","ref_index":39,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VHHKA3XC6YF3JDBVZ7NI2JJ62U","json":"https://pith.science/pith/VHHKA3XC6YF3JDBVZ7NI2JJ62U.json","graph_json":"https://pith.science/api/pith-number/VHHKA3XC6YF3JDBVZ7NI2JJ62U/graph.json","events_json":"https://pith.science/api/pith-number/VHHKA3XC6YF3JDBVZ7NI2JJ62U/events.json","paper":"https://pith.science/paper/VHHKA3XC"},"agent_actions":{"view_html":"https://pith.science/pith/VHHKA3XC6YF3JDBVZ7NI2JJ62U","download_json":"https://pith.science/pith/VHHKA3XC6YF3JDBVZ7NI2JJ62U.json","view_paper":"https://pith.science/paper/VHHKA3XC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2005.00850&json=true","fetch_graph":"https://pith.science/api/pith-number/VHHKA3XC6YF3JDBVZ7NI2JJ62U/graph.json","fetch_events":"https://pith.science/api/pith-number/VHHKA3XC6YF3JDBVZ7NI2JJ62U/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VHHKA3XC6YF3JDBVZ7NI2JJ62U/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VHHKA3XC6YF3JDBVZ7NI2JJ62U/action/storage_attestation","attest_author":"https://pith.science/pith/VHHKA3XC6YF3JDBVZ7NI2JJ62U/action/author_attestation","sign_citation":"https://pith.science/pith/VHHKA3XC6YF3JDBVZ7NI2JJ62U/action/citation_signature","submit_replication":"https://pith.science/pith/VHHKA3XC6YF3JDBVZ7NI2JJ62U/action/replication_record"}},"created_at":"2026-07-05T01:02:30.145652+00:00","updated_at":"2026-07-05T01:02:30.145652+00:00"}