{"id":"https://openalex.org/W4412887910","doi":"https://doi.org/10.18653/v1/2025.findings-acl.1088","title":"Implicit Cross-Lingual Rewarding for Efficient Multilingual Preference Alignment","display_name":"Implicit Cross-Lingual Rewarding for Efficient Multilingual Preference Alignment","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412887910","doi":"https://doi.org/10.18653/v1/2025.findings-acl.1088"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-acl.1088","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1088","pdf_url":"https://aclanthology.org/2025.findings-acl.1088.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.1088.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5069540177","display_name":"Wen Yang","orcid":"https://orcid.org/0000-0002-3263-8768"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wen Yang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","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/A5100337673","display_name":"Chen Wang","orcid":"https://orcid.org/0000-0003-2922-5345"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","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":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5100319572","display_name":"Jiajun Zhang","orcid":"https://orcid.org/0000-0001-5293-7434"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiajun Zhang","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":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"21125","last_page":"21147"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12970","display_name":"Names, Identity, and Discrimination Research","score":0.7764000296592712,"subfield":{"id":"https://openalex.org/subfields/3312","display_name":"Sociology and Political Science"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T12970","display_name":"Names, Identity, and Discrimination Research","score":0.7764000296592712,"subfield":{"id":"https://openalex.org/subfields/3312","display_name":"Sociology and Political Science"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T10991","display_name":"Game Theory and Voting Systems","score":0.7501999735832214,"subfield":{"id":"https://openalex.org/subfields/2002","display_name":"Economics and Econometrics"},"field":{"id":"https://openalex.org/fields/20","display_name":"Economics, Econometrics and Finance"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T10203","display_name":"Recommender Systems and Techniques","score":0.6780999898910522,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/computer-science","display_name":"Computer science","score":0.7111340165138245},{"id":"https://openalex.org/keywords/preference","display_name":"Preference","score":0.5924080014228821},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.476109117269516},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.3976309299468994},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.19061508774757385},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.10790634155273438}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7111340165138245},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.5924080014228821},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.476109117269516},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3976309299468994},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.19061508774757385},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.10790634155273438}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-acl.1088","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1088","pdf_url":"https://aclanthology.org/2025.findings-acl.1088.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.1088","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1088","pdf_url":"https://aclanthology.org/2025.findings-acl.1088.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":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320321133","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412887910.pdf","grobid_xml":"https://content.openalex.org/works/W4412887910.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W3204019825"],"abstract_inverted_index":{"Direct":[0],"Preference":[1],"Optimization":[2],"(DPO)":[3],"has":[4,19],"become":[5],"a":[6,39,125,133],"prominent":[7],"method":[8],"for":[9,106,120,168],"aligning":[10,24],"Large":[11],"Language":[12],"Models":[13],"(LLMs)":[14],"with":[15],"human":[16],"preferences.While":[17],"DPO":[18,108],"enabled":[20],"significant":[21],"progress":[22],"in":[23,90,124,129,136],"English":[25,48,73,95,115],"LLMs,":[26],"multilingual":[27,99,107,161,170],"preference":[28,88,111,162,171],"alignment":[29],"is":[30,83,103,174],"hampered":[31],"by":[32,50],"data":[33,102],"scarcity.To":[34],"address":[35],"this,":[36],"we":[37,62],"propose":[38],"novel":[40],"approach":[41],"that":[42,151],"captures":[43],"learned":[44],"preferences":[45],"from":[46,68,114],"well-aligned":[47],"models":[49,155],"implicit":[51,65],"rewards":[52],"and":[53,76,132,159],"transfers":[54],"them":[55],"to":[56,86,97,116],"other":[57,117],"languages":[58,144],"through":[59],"iterative":[60],"training.Specifically,":[61],"derive":[63],"an":[64,72],"reward":[66,81],"model":[67,75,82],"the":[69,146,166],"logits":[70],"of":[71],"DPO-aligned":[74],"its":[77],"corresponding":[78],"reference":[79],"model.This":[80],"then":[84],"leveraged":[85],"annotate":[87],"relations":[89],"cross-lingual":[91],"instructionresponse":[92],"pairs,":[93],"using":[94],"instructions":[96],"evaluate":[98],"responses.The":[100],"annotated":[101],"subsequently":[104],"used":[105],"fine-tuning,":[109],"facilitating":[110],"knowledge":[112],"transfer":[113],"languages.Fine-tuning":[118],"Llama3":[119],"two":[121],"iterations":[122],"resulted":[123],"12.72%":[126],"average":[127],"improvement":[128],"Win":[130,139],"Rate":[131,140],"5.97%":[134],"increase":[135],"Length":[137],"Control":[138],"across":[141],"all":[142],"training":[143],"on":[145],"X-AlpacaEval":[147],"leaderboard.Our":[148],"findings":[149],"demonstrate":[150],"leveraging":[152],"existing":[153],"English-aligned":[154],"can":[156],"enable":[157],"efficient":[158],"effective":[160],"alignment,":[163],"significantly":[164],"reducing":[165],"need":[167],"extensive":[169],"data.The":[172],"code":[173],"available":[175],"at":[176],"https://github.com/ZNLP/":[177],"Implicit-Cross-Lingual-Rewarding.":[178]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}