{"id":"https://openalex.org/W4414588048","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1685","title":"Curriculum-RLAIF: Curriculum Alignment with Reinforcement Learning from AI Feedback","display_name":"Curriculum-RLAIF: Curriculum Alignment with Reinforcement Learning from AI Feedback","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W4414588048","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1685"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.1685","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1685","pdf_url":"https://aclanthology.org/2026.findings-acl.1685.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":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.findings-acl.1685.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100606513","display_name":"Mengdi Li","orcid":"https://orcid.org/0009-0000-2650-2891"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiaye Lin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Mengdi Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mengdi Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103057571","display_name":"Xufeng Zhao","orcid":"https://orcid.org/0009-0001-6972-5048"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xufeng Zhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023237216","display_name":"Wenhao Lu","orcid":"https://orcid.org/0000-0001-8660-2886"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wenhao Lu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015991234","display_name":"Peilin Zhao","orcid":"https://orcid.org/0000-0001-8543-3953"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Peilin Zhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5033486668","display_name":"Stefan Wermter","orcid":"https://orcid.org/0000-0003-1343-4775"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Stefan Wermter","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5075652117","display_name":"Di Wang","orcid":"https://orcid.org/0000-0002-7732-1035"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Di Wang","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":0.00675604,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"33757","last_page":"33777"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10906","display_name":"AI-based Problem Solving and Planning","score":0.462799996137619,"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/T10906","display_name":"AI-based Problem Solving and Planning","score":0.462799996137619,"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/T12423","display_name":"Software Reliability and Analysis Research","score":0.45910000801086426,"subfield":{"id":"https://openalex.org/subfields/1712","display_name":"Software"},"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":0.7462000250816345},{"id":"https://openalex.org/keywords/margin","display_name":"Margin (machine learning)","score":0.6753000020980835},{"id":"https://openalex.org/keywords/generalizability-theory","display_name":"Generalizability theory","score":0.632099986076355},{"id":"https://openalex.org/keywords/preference","display_name":"Preference","score":0.5976999998092651},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5044000148773193},{"id":"https://openalex.org/keywords/perspective","display_name":"Perspective (graphical)","score":0.4964999854564667},{"id":"https://openalex.org/keywords/selection","display_name":"Selection (genetic algorithm)","score":0.4194999933242798},{"id":"https://openalex.org/keywords/preference-learning","display_name":"Preference learning","score":0.40639999508857727}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7462000250816345},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6801999807357788},{"id":"https://openalex.org/C774472","wikidata":"https://www.wikidata.org/wiki/Q6760393","display_name":"Margin (machine learning)","level":2,"score":0.6753000020980835},{"id":"https://openalex.org/C27158222","wikidata":"https://www.wikidata.org/wiki/Q5532422","display_name":"Generalizability theory","level":2,"score":0.632099986076355},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6265000104904175},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.6025999784469604},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.5976999998092651},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5044000148773193},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.4964999854564667},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.4194999933242798},{"id":"https://openalex.org/C181204326","wikidata":"https://www.wikidata.org/wiki/Q7239820","display_name":"Preference learning","level":3,"score":0.40639999508857727},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.35010001063346863},{"id":"https://openalex.org/C2777868144","wikidata":"https://www.wikidata.org/wiki/Q7239817","display_name":"Preference elicitation","level":3,"score":0.349700003862381},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.3458999991416931},{"id":"https://openalex.org/C47177190","wikidata":"https://www.wikidata.org/wiki/Q207137","display_name":"Curriculum","level":2,"score":0.3206000030040741},{"id":"https://openalex.org/C2779436431","wikidata":"https://www.wikidata.org/wiki/Q30672407","display_name":"Policy learning","level":2,"score":0.3149000108242035},{"id":"https://openalex.org/C93959086","wikidata":"https://www.wikidata.org/wiki/Q6888345","display_name":"Model selection","level":2,"score":0.30219998955726624},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.2596000134944916},{"id":"https://openalex.org/C2779955035","wikidata":"https://www.wikidata.org/wiki/Q4686785","display_name":"Advice (programming)","level":2,"score":0.25540000200271606},{"id":"https://openalex.org/C160234255","wikidata":"https://www.wikidata.org/wiki/Q812535","display_name":"Bayesian inference","level":3,"score":0.2508000135421753}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.1685","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1685","pdf_url":"https://aclanthology.org/2026.findings-acl.1685.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"},{"id":"pmh:oai:arXiv.org:2505.20075","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2505.20075","pdf_url":"https://arxiv.org/pdf/2505.20075","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"publisher-specific-oa","license_id":"https://openalex.org/licenses/publisher-specific-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"pmh:oai:arXiv.org:2505.20075","is_oa":true,"landing_page_url":"https://arxiv.org/abs/2505.20075","pdf_url":"https://arxiv.org/pdf/2505.20075","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":"publisher-specific-oa","license_id":"https://openalex.org/licenses/publisher-specific-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2505.20075","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2505.20075","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":"Preprint"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-acl.1685","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1685","pdf_url":"https://aclanthology.org/2026.findings-acl.1685.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":[{"id":"https://openalex.org/G4795079661","display_name":null,"funder_award_id":"BAS/1/1689-01-01","funder_id":"https://openalex.org/F4320322320","funder_display_name":"King Abdullah University of Science and Technology"}],"funders":[{"id":"https://openalex.org/F4320322320","display_name":"King Abdullah University of Science and Technology","ror":"https://ror.org/01q3tbs38"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4414588048.pdf","grobid_xml":"https://content.openalex.org/works/W4414588048.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Reward":[0],"models":[1,54,105,118],"trained":[2,106],"through":[3,55],"Reinforcement":[4],"Learning":[5],"from":[6,13,26,69],"AI":[7],"Feedback":[8],"(RLAIF)":[9],"methods":[10],"frequently":[11],"suffer":[12],"limited":[14],"generalizability,":[15,111],"which":[16,82],"hinders":[17],"the":[18,50,61,113,141],"alignment":[19,114],"performance":[20,115],"of":[21,37,52,73,116,143],"policy":[22,117],"models.This":[23],"challenge":[24],"stems":[25],"various":[27,130],"issues,":[28],"including":[29],"distribution":[30],"shift,":[31],"preference":[32,84],"label":[33],"noise,":[34],"and":[35,90,135,148],"mismatch":[36],"overly":[38],"challenging":[39],"samples":[40],"with":[41,86,107,137],"model":[42,98],"capacity.In":[43],"this":[44],"paper,":[45],"we":[46,76],"aim":[47],"to":[48,129],"enhance":[49],"generalizability":[51],"reward":[53,97,104],"a":[56,70,78,93,120],"data-centric":[57],"approach,":[58],"driven":[59],"by":[60,119],"insight":[62],"that":[63,103],"these":[64],"issues":[65],"are":[66],"inherently":[67],"intertwined":[68],"uniform":[71],"perspective":[72],"data":[74],"difficulty.Accordingly,":[75],"propose":[77],"novel":[79],"framework,":[80],"Curriculum-RLAIF,":[81],"constructs":[83],"pairs":[85],"varying":[87],"difficulty":[88],"levels":[89],"then":[91],"produces":[92],"specific":[94],"curriculum":[95],"for":[96],"training.Comprehensive":[99],"experimental":[100],"results":[101],"suggest":[102],"Curriculum-RLAIF":[108,144],"achieve":[109],"improved":[110],"boosting":[112],"significant":[121],"margin":[122],"without":[123],"incurring":[124],"additional":[125],"inference":[126],"costs":[127],"compared":[128],"existing":[131],"non-curriculum":[132],"baselines.Further":[133],"analysis":[134],"comparison":[136],"alternative":[138],"strategies":[139],"highlight":[140],"superiority":[142],"in":[145],"simplicity,":[146],"efficiency,":[147],"effectiveness.":[149]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}