{"id":"https://openalex.org/W4412888237","doi":"https://doi.org/10.18653/v1/2025.findings-acl.632","title":"Boosting LLM Translation Skills without General Ability Loss via Rationale Distillation","display_name":"Boosting LLM Translation Skills without General Ability Loss via Rationale Distillation","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412888237","doi":"https://doi.org/10.18653/v1/2025.findings-acl.632"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-acl.632","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.632","pdf_url":"https://aclanthology.org/2025.findings-acl.632.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 2025","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-acl.632.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101499870","display_name":"Junhong Wu","orcid":"https://orcid.org/0000-0002-2679-1627"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Junhong Wu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5056718303","display_name":"Yang Zhao","orcid":"https://orcid.org/0000-0002-7302-7787"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang Zhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5055186731","display_name":"Yangyifan Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yangyifan Xu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5088849506","display_name":"Bing Liu","orcid":"https://orcid.org/0009-0006-4233-0667"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bing Liu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5015785439","display_name":"Chengqing Zong","orcid":"https://orcid.org/0000-0002-9864-3818"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chengqing Zong","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":null,"has_fulltext":true,"cited_by_count":2,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"12217","last_page":"12236"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9944999814033508,"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":0.9944999814033508,"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/T13629","display_name":"Text Readability and Simplification","score":0.9205999970436096,"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/boosting","display_name":"Boosting (machine learning)","score":0.8095272779464722},{"id":"https://openalex.org/keywords/distillation","display_name":"Distillation","score":0.596703052520752},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5876704454421997},{"id":"https://openalex.org/keywords/translation","display_name":"Translation (biology)","score":0.528944730758667},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5147738456726074},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.4534100294113159},{"id":"https://openalex.org/keywords/chemistry","display_name":"Chemistry","score":0.15801012516021729},{"id":"https://openalex.org/keywords/chromatography","display_name":"Chromatography","score":0.07265132665634155}],"concepts":[{"id":"https://openalex.org/C46686674","wikidata":"https://www.wikidata.org/wiki/Q466303","display_name":"Boosting (machine learning)","level":2,"score":0.8095272779464722},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.596703052520752},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5876704454421997},{"id":"https://openalex.org/C149364088","wikidata":"https://www.wikidata.org/wiki/Q185917","display_name":"Translation (biology)","level":4,"score":0.528944730758667},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5147738456726074},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4534100294113159},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.15801012516021729},{"id":"https://openalex.org/C43617362","wikidata":"https://www.wikidata.org/wiki/Q170050","display_name":"Chromatography","level":1,"score":0.07265132665634155},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"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/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-acl.632","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.632","pdf_url":"https://aclanthology.org/2025.findings-acl.632.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 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-acl.632","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.632","pdf_url":"https://aclanthology.org/2025.findings-acl.632.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 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[{"display_name":"Quality Education","score":0.7799999713897705,"id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412888237.pdf","grobid_xml":"https://content.openalex.org/works/W4412888237.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2961085424","https://openalex.org/W4306674287","https://openalex.org/W3082059448","https://openalex.org/W4313640622","https://openalex.org/W4387369504","https://openalex.org/W4394896187","https://openalex.org/W3170094116","https://openalex.org/W4386462264","https://openalex.org/W3107602296","https://openalex.org/W4364306694"],"abstract_inverted_index":{"Large":[0],"Language":[1],"Models":[2],"(LLMs)":[3],"have":[4],"achieved":[5],"impressive":[6],"results":[7],"across":[8,126],"numerous":[9],"NLP":[10],"tasks,":[11,96],"and":[12,37,48,111,141],"fine-tuning":[13,24],"them":[14],"for":[15,80,153],"Machine":[16],"Translation":[17],"(MT)":[18],"has":[19,142],"improved":[20],"their":[21],"performance.However,":[22],"vanilla":[23],"often":[25],"leads":[26],"to":[27,77,87,101,145],"catastrophic":[28],"forgetting,":[29],"compromising":[30],"the":[31,71,103,114,138,143],"broad":[32],"general":[33,124,149],"abilities":[34,125],"of":[35,75,156],"LLMs":[36,76],"introducing":[38],"potential":[39,144],"security":[40],"risks.These":[41],"abilities,":[42],"which":[43,83],"are":[44,84],"developed":[45],"using":[46,135],"proprietary":[47],"unavailable":[49],"training":[50,81,104,107],"data,":[51,82],"make":[52],"simple":[53],"data":[54],"replay":[55],"methods":[56],"ineffective.To":[57],"overcome":[58],"this":[59],"issue,":[60],"we":[61],"propose":[62],"a":[63,131,148,154],"novel":[64],"approach":[65],"called":[66],"RaDis":[67,129],"(Rationale":[68],"Distillation).RaDis":[69],"harnesses":[70],"strong":[72],"generative":[73],"capabilities":[74],"create":[78],"rationales":[79,90,136],"then":[85],"\"replayed\"":[86],"prevent":[88],"forgetting.These":[89],"connect":[91],"prior":[92],"knowledge":[93],"with":[94],"new":[95,118],"acting":[97],"as":[98,147],"selfdistillation":[99],"targets":[100],"regulate":[102],"process.By":[105],"jointly":[106],"on":[108,134],"reference":[109],"translations":[110],"self-generated":[112],"rationales,":[113],"model":[115],"can":[116],"learn":[117],"translation":[119],"skills":[120],"while":[121],"preserving":[122],"its":[123],"other":[127],"tasks.Additionally,":[128],"provides":[130],"fresh":[132],"perspective":[133],"in":[137],"CL":[139],"field":[140],"serve":[146],"continual":[150],"learning":[151],"method":[152],"variety":[155],"tasks.":[157]},"counts_by_year":[{"year":2025,"cited_by_count":2}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}