{"id":"https://openalex.org/W7166880276","doi":"https://doi.org/10.18653/v1/2026.findings-acl.851","title":"PEGRL: Improving Machine Translation by Post-Editing Guided Reinforcement Learning","display_name":"PEGRL: Improving Machine Translation by Post-Editing Guided Reinforcement Learning","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166880276","doi":"https://doi.org/10.18653/v1/2026.findings-acl.851"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.851","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.851","pdf_url":"https://aclanthology.org/2026.findings-acl.851.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.findings-acl.851.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5110913682","display_name":"Yunzhi Shen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yunzhi Shen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139732108","display_name":"Hao Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hao Zhou","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139715143","display_name":"Xin Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xin Huang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139763429","display_name":"Xue Han","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xue Han","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139749481","display_name":"Junlan Feng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Junlan Feng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5139735411","display_name":"Shujian Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shujian Huang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.86854985,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"17225","last_page":"17242"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.7336999773979187,"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":0.7336999773979187,"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":0.028300000354647636,"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"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.027699999511241913,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.4652000069618225},{"id":"https://openalex.org/keywords/machine-translation","display_name":"Machine translation","score":0.359499990940094},{"id":"https://openalex.org/keywords/translation","display_name":"Translation (biology)","score":0.3084000051021576},{"id":"https://openalex.org/keywords/control","display_name":"Control (management)","score":0.2816999852657318},{"id":"https://openalex.org/keywords/active-learning","display_name":"Active learning (machine learning)","score":0.2802000045776367}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6636000275611877},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5630000233650208},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.4652000069618225},{"id":"https://openalex.org/C203005215","wikidata":"https://www.wikidata.org/wiki/Q79798","display_name":"Machine translation","level":2,"score":0.359499990940094},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3569999933242798},{"id":"https://openalex.org/C149364088","wikidata":"https://www.wikidata.org/wiki/Q185917","display_name":"Translation (biology)","level":4,"score":0.3084000051021576},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2896000146865845},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.2816999852657318},{"id":"https://openalex.org/C77967617","wikidata":"https://www.wikidata.org/wiki/Q4677561","display_name":"Active learning (machine learning)","level":2,"score":0.2802000045776367},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.2703000009059906},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.26930001378059387},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.25360000133514404}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.851","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.851","pdf_url":"https://aclanthology.org/2026.findings-acl.851.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-acl.851","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.851","pdf_url":"https://aclanthology.org/2026.findings-acl.851.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166880276.pdf","grobid_xml":"https://content.openalex.org/works/W7166880276.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Reinforcement":[0],"learning":[1,27],"(RL)":[2],"has":[3],"shown":[4],"strong":[5],"promise":[6],"for":[7,137],"LLM-based":[8,146],"machine":[9],"translation,":[10],"with":[11],"recent":[12],"methods":[13],"such":[14],"as":[15,35,37,59],"GRPO":[16],"demonstrating":[17],"notable":[18],"gains;":[19],"nevertheless,":[20],"translationoriented":[21],"RL":[22,54,134],"remains":[23],"challenged":[24],"by":[25],"noisy":[26],"signals":[28],"arising":[29],"from":[30,89],"Monte":[31],"Carlo":[32],"return":[33,81],"estimation,":[34],"well":[36],"a":[38,52,119,151],"large":[39],"trajectory":[40],"space":[41],"that":[42,56],"favors":[43],"global":[44,100],"exploration":[45,101],"over":[46,133],"fine-grained":[47,103],"local":[48,104],"optimization.We":[49],"introduce":[50],"PEGRL,":[51],"twostage":[53],"framework":[55],"uses":[57],"post-editing":[58,85,116],"an":[60],"auxiliary":[61],"task":[62],"to":[63,76,87,144],"stabilize":[64],"training":[65],"and":[66,102,115,128,136,150],"guide":[67],"overall":[68],"optimization.At":[69],"each":[70],"iteration,":[71],"translation":[72,94,114],"outputs":[73],"are":[74,157],"sampled":[75],"construct":[77],"postediting":[78],"inputs,":[79],"allowing":[80],"estimation":[82],"in":[83],"the":[84,92,111],"stage":[86],"benefit":[88],"conditioning":[90],"on":[91,125,140],"current":[93],"behavior,":[95],"while":[96],"jointly":[97],"supporting":[98],"both":[99],"optimization.A":[105],"task-specific":[106],"weighting":[107],"scheme":[108],"further":[109],"balances":[110],"contributions":[112],"of":[113,153],"objectives,":[117],"yielding":[118],"biased":[120],"yet":[121],"more":[122],"sample-efficient":[123],"estimator.Experiments":[124],"English\u2192Finnish,":[126],"English\u2192Turkish,":[127,138],"English\u2194Chinese":[129],"show":[130],"consistent":[131],"gains":[132],"baselines,":[135],"performance":[139],"COMET-KIWI":[141],"is":[142],"comparable":[143],"advanced":[145],"systems":[147],"(DeepSeek-V3.2).Our":[148],"code":[149],"set":[152],"representative":[154],"pretrained":[155],"models":[156],"publicly":[158],"available":[159],"at":[160],"https:":[161]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}