{"id":"https://openalex.org/W7140132325","doi":"https://doi.org/10.18653/v1/2026.findings-eacl.99","title":"Best-of-L: Cross-Lingual Reward Modeling for Mathematical Reasoning","display_name":"Best-of-L: Cross-Lingual Reward Modeling for Mathematical Reasoning","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7140132325","doi":"https://doi.org/10.18653/v1/2026.findings-eacl.99"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.findings-eacl.99","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-eacl.99","pdf_url":"https://aclanthology.org/2026.findings-eacl.99.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: EACL 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-eacl.99.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5130344851","display_name":"Sara Rajaee","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sara Rajaee","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109855717","display_name":"Rochelle Choenni","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rochelle Choenni","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130371258","display_name":"Ekaterina Shutova","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ekaterina Shutova","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130404962","display_name":"Christof Monz","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Christof Monz","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":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1930","last_page":"1939"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.2093999981880188,"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/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.2093999981880188,"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/T11596","display_name":"Constraint Satisfaction and Optimization","score":0.17350000143051147,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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":0.12470000237226486,"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/mathematical-model","display_name":"Mathematical model","score":0.3540000021457672},{"id":"https://openalex.org/keywords/automated-reasoning","display_name":"Automated reasoning","score":0.30630001425743103},{"id":"https://openalex.org/keywords/model-based-reasoning","display_name":"Model-based reasoning","score":0.29789999127388},{"id":"https://openalex.org/keywords/non-monotonic-logic","display_name":"Non-monotonic logic","score":0.2766000032424927},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.24140000343322754}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5873000025749207},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5141000151634216},{"id":"https://openalex.org/C76969082","wikidata":"https://www.wikidata.org/wiki/Q486902","display_name":"Mathematical model","level":2,"score":0.3540000021457672},{"id":"https://openalex.org/C195344581","wikidata":"https://www.wikidata.org/wiki/Q2555318","display_name":"Automated reasoning","level":2,"score":0.30630001425743103},{"id":"https://openalex.org/C37335422","wikidata":"https://www.wikidata.org/wiki/Q6888134","display_name":"Model-based reasoning","level":3,"score":0.29789999127388},{"id":"https://openalex.org/C159032336","wikidata":"https://www.wikidata.org/wiki/Q2488768","display_name":"Non-monotonic logic","level":2,"score":0.2766000032424927},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.25780001282691956},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.24140000343322754},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.23980000615119934},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.23919999599456787}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-eacl.99","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-eacl.99","pdf_url":"https://aclanthology.org/2026.findings-eacl.99.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: EACL 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-eacl.99","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-eacl.99","pdf_url":"https://aclanthology.org/2026.findings-eacl.99.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: EACL 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[{"display_name":"Quality Education","score":0.46515968441963196,"id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7140132325.pdf","grobid_xml":"https://content.openalex.org/works/W7140132325.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"While":[0],"the":[1,72,113,127],"reasoning":[2,30,37,60,124],"abilities":[3,17],"of":[4,130],"large":[5],"language":[6],"models":[7],"(LLMs)":[8],"continue":[9],"to":[10,84,88,121],"advance,":[11],"it":[12],"remains":[13],"underexplored":[14],"how":[15],"such":[16],"vary":[18],"across":[19,62],"languages":[20,27,42,63],"in":[21,40],"multilingual":[22,123],"LLMs":[23],"and":[24],"whether":[25],"different":[26,41],"generate":[28],"distinct":[29],"paths.In":[31],"this":[32],"work,":[33],"we":[34],"show":[35,75],"that":[36,57,76],"traces":[38,61],"generated":[39],"often":[43],"provide":[44],"complementary":[45,128],"signals":[46],"for":[47],"mathematical":[48],"reasoning.We":[49],"propose":[50],"cross-lingual":[51,77,101],"outcome":[52],"reward":[53,78,90],"modeling,":[54],"a":[55,67,93],"framework":[56],"ranks":[58],"candidate":[59],"rather":[64],"than":[65],"within":[66,92],"single":[68,94],"language.Our":[69,116],"experiments":[70],"on":[71],"MGSM":[73],"benchmark":[74],"modeling":[79,91],"improves":[80,103],"accuracy":[81],"by":[82,125],"up":[83],"10":[85],"points":[86],"compared":[87],"using":[89],"language,":[95],"benefiting":[96],"both":[97],"high-and":[98],"low-resource":[99],"languages.Notably,":[100],"sampling":[102],"English":[104,111],"performance":[105],"under":[106],"low":[107],"inference":[108],"budgets,":[109],"despite":[110],"being":[112],"strongest":[114],"individual":[115],"findings":[117],"reveal":[118],"new":[119],"opportunities":[120],"improve":[122],"leveraging":[126],"strengths":[129],"diverse":[131],"languages.":[132]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2026-03-24T00:00:00"}