{"id":"https://openalex.org/W2460130460","doi":"https://doi.org/10.18653/v1/k16-1029","title":"Compression of Neural Machine Translation Models via Pruning","display_name":"Compression of Neural Machine Translation Models via Pruning","publication_year":2016,"publication_date":"2016-01-01","ids":{"openalex":"https://openalex.org/W2460130460","doi":"https://doi.org/10.18653/v1/k16-1029","mag":"2460130460"},"language":"en","primary_location":{"id":"doi:10.18653/v1/k16-1029","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/k16-1029","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of The 20th SIGNLL Conference on Computational Natural Language Learning","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.18653/v1/k16-1029","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5076818344","display_name":"Abigail See","orcid":"https://orcid.org/0000-0003-3137-6599"},"institutions":[{"id":"https://openalex.org/I241749","display_name":"University of Cambridge","ror":"https://ror.org/013meh722","country_code":"GB","type":"education","lineage":["https://openalex.org/I241749"]}],"countries":["GB"],"is_corresponding":true,"raw_author_name":"Abigail See","raw_affiliation_strings":["University of Cambridge, Cambridge, United Kingdom"],"affiliations":[{"raw_affiliation_string":"University of Cambridge, Cambridge, United Kingdom","institution_ids":["https://openalex.org/I241749"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078684067","display_name":"Minh-Thang Luong","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Minh-Thang Luong","raw_affiliation_strings":["Google (United States), Mountain View, United States"],"affiliations":[{"raw_affiliation_string":"Google (United States), Mountain View, United States","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5046006076","display_name":"Christopher D. Manning","orcid":"https://orcid.org/0000-0001-6155-649X"},"institutions":[{"id":"https://openalex.org/I97018004","display_name":"Stanford University","ror":"https://ror.org/00f54p054","country_code":"US","type":"education","lineage":["https://openalex.org/I97018004"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Christopher D. Manning","raw_affiliation_strings":["Stanford University, Stanford, United States"],"affiliations":[{"raw_affiliation_string":"Stanford University, Stanford, United States","institution_ids":["https://openalex.org/I97018004"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5076818344"],"corresponding_institution_ids":["https://openalex.org/I241749"],"apc_list":null,"apc_paid":null,"fwci":6.20129134,"has_fulltext":false,"cited_by_count":50,"citation_normalized_percentile":{"value":0.98324884,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"291","last_page":"301"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9997000098228455,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9976999759674072,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.7533027529716492},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7521449327468872},{"id":"https://openalex.org/keywords/machine-translation","display_name":"Machine translation","score":0.7258576154708862},{"id":"https://openalex.org/keywords/redundancy","display_name":"Redundancy (engineering)","score":0.6248590350151062},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5070703029632568},{"id":"https://openalex.org/keywords/translation","display_name":"Translation (biology)","score":0.4845813512802124},{"id":"https://openalex.org/keywords/architecture","display_name":"Architecture","score":0.42017853260040283}],"concepts":[{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.7533027529716492},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7521449327468872},{"id":"https://openalex.org/C203005215","wikidata":"https://www.wikidata.org/wiki/Q79798","display_name":"Machine translation","level":2,"score":0.7258576154708862},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.6248590350151062},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5070703029632568},{"id":"https://openalex.org/C149364088","wikidata":"https://www.wikidata.org/wiki/Q185917","display_name":"Translation (biology)","level":4,"score":0.4845813512802124},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.42017853260040283},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C153349607","wikidata":"https://www.wikidata.org/wiki/Q36649","display_name":"Visual arts","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C6557445","wikidata":"https://www.wikidata.org/wiki/Q173113","display_name":"Agronomy","level":1,"score":0.0},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0},{"id":"https://openalex.org/C142362112","wikidata":"https://www.wikidata.org/wiki/Q735","display_name":"Art","level":0,"score":0.0},{"id":"https://openalex.org/C105580179","wikidata":"https://www.wikidata.org/wiki/Q188928","display_name":"Messenger RNA","level":3,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.18653/v1/k16-1029","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/k16-1029","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of The 20th SIGNLL Conference on Computational Natural Language Learning","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:1606.09274","is_oa":true,"landing_page_url":"http://arxiv.org/abs/1606.09274","pdf_url":"https://arxiv.org/pdf/1606.09274","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},{"id":"mag:2460130460","is_oa":true,"landing_page_url":"http://export.arxiv.org/pdf/1606.09274","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null},{"id":"doi:10.48550/arxiv.1606.09274","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.1606.09274","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.18653/v1/k16-1029","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/k16-1029","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of The 20th SIGNLL Conference on Computational Natural Language Learning","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":27,"referenced_works":["https://openalex.org/W1570197553","https://openalex.org/W1591801644","https://openalex.org/W1753482797","https://openalex.org/W1800356822","https://openalex.org/W1821462560","https://openalex.org/W1825672851","https://openalex.org/W1841592590","https://openalex.org/W1902237438","https://openalex.org/W2064675550","https://openalex.org/W2114766824","https://openalex.org/W2125389748","https://openalex.org/W2167215970","https://openalex.org/W2184135559","https://openalex.org/W2279098554","https://openalex.org/W2337978879","https://openalex.org/W2339995566","https://openalex.org/W2949888546","https://openalex.org/W2950967261","https://openalex.org/W2952432176","https://openalex.org/W2952533036","https://openalex.org/W2953071172","https://openalex.org/W2963216553","https://openalex.org/W2963674932","https://openalex.org/W2964200805","https://openalex.org/W2964299589","https://openalex.org/W2964308564","https://openalex.org/W3204406378"],"related_works":["https://openalex.org/W2963674932","https://openalex.org/W2114766824","https://openalex.org/W2964299589","https://openalex.org/W1821462560","https://openalex.org/W2194775991","https://openalex.org/W2962784628","https://openalex.org/W2612445135","https://openalex.org/W2608554408","https://openalex.org/W2125389748","https://openalex.org/W2963403868","https://openalex.org/W2525778437","https://openalex.org/W2964308564","https://openalex.org/W2963341956","https://openalex.org/W2515385951","https://openalex.org/W1522301498","https://openalex.org/W2513419314","https://openalex.org/W2251939518","https://openalex.org/W1724438581","https://openalex.org/W2095705004","https://openalex.org/W2064675550"],"abstract_inverted_index":{"Neural":[0],"Machine":[1],"Translation":[2],"(NMT),":[3],"like":[4],"many":[5],"other":[6],"deep":[7],"learning":[8],"domains,":[9],"typically":[10],"suffers":[11],"from":[12],"over-parameterization,":[13],"resulting":[14],"in":[15,38,52,109],"large":[16],"storage":[17],"sizes.":[18],"This":[19,101],"paper":[20],"examines":[21],"three":[22],"simple":[23],"magnitude-based":[24],"pruning":[25,42,62],"schemes":[26],"to":[27],"compress":[28],"NMT":[29,54,70,76,111],"models,":[30],"namely":[31],"class-blind,":[32],"class-uniform,":[33],"and":[34,123],"class-distribution,":[35],"which":[36],"differ":[37],"terms":[39],"of":[40,50,60,107],"how":[41],"thresholds":[43],"are":[44],"computed":[45],"for":[46,67],"the":[47,53,58,96,105,110,126],"different":[48],"classes":[49],"weights":[51],"architecture.":[55,112],"We":[56,72],"demonstrate":[57],"efficacy":[59],"weight":[61],"as":[63,93],"a":[64,68],"compression":[65],"technique":[66],"state-of-the-art":[69],"system.":[71],"show":[73],"that":[74,117],"an":[75,130],"model":[77],"with":[78,88,118,129],"over":[79],"200":[80],"million":[81],"parameters":[82],"can":[83,121],"be":[84],"pruned":[85],"by":[86],"40%":[87],"very":[89],"little":[90],"performance":[91,128],"loss":[92],"measured":[94],"on":[95,104],"WMT'14":[97],"English-German":[98],"translation":[99],"task.":[100],"sheds":[102],"light":[103],"distribution":[106],"redundancy":[108],"Our":[113],"main":[114],"result":[115],"is":[116],"retraining,":[119],"we":[120],"recover":[122],"even":[124],"surpass":[125],"original":[127],"80%-pruned":[131],"model.":[132]},"counts_by_year":[{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":3},{"year":2021,"cited_by_count":8},{"year":2020,"cited_by_count":13},{"year":2019,"cited_by_count":13},{"year":2018,"cited_by_count":4},{"year":2017,"cited_by_count":3},{"year":2016,"cited_by_count":2}],"updated_date":"2026-02-09T05:59:30.833894","created_date":"2025-10-10T00:00:00"}