{"id":"https://openalex.org/W4416034889","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.412","title":"Scale Down to Speed Up: Dynamic Data Selection for Reinforcement Learning","display_name":"Scale Down to Speed Up: Dynamic Data Selection for Reinforcement Learning","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416034889","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.412"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.412","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.412","pdf_url":"https://aclanthology.org/2025.findings-emnlp.412.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: EMNLP 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-emnlp.412.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5006569196","display_name":"Zhuoyue Chen","orcid":"https://orcid.org/0000-0001-9585-5607"},"institutions":[{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhuoyue Chen","raw_affiliation_strings":["Zhejiang University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang University","institution_ids":["https://openalex.org/I76130692"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101751726","display_name":"Jihai Zhang","orcid":"https://orcid.org/0000-0003-4242-5341"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jihai Zhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082502751","display_name":"Ben Liu","orcid":"https://orcid.org/0000-0001-5031-9368"},"institutions":[{"id":"https://openalex.org/I37461747","display_name":"Wuhan University","ror":"https://ror.org/033vjfk17","country_code":"CN","type":"education","lineage":["https://openalex.org/I37461747"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ben Liu","raw_affiliation_strings":["Wuhan University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Wuhan University","institution_ids":["https://openalex.org/I37461747"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5086363516","display_name":"Fangquan Lin","orcid":"https://orcid.org/0000-0001-5341-378X"},"institutions":[{"id":"https://openalex.org/I4210086143","display_name":"Alibaba Group (Cayman Islands)","ror":"https://ror.org/00mnrxf72","country_code":"KY","type":"company","lineage":["https://openalex.org/I4210086143","https://openalex.org/I45928872"]}],"countries":["KY"],"is_corresponding":false,"raw_author_name":"Fangquan Lin","raw_affiliation_strings":["DAMO Academy , Alibaba group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"DAMO Academy , Alibaba group","institution_ids":["https://openalex.org/I4210086143"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5109746529","display_name":"Wotao Yin","orcid":null},"institutions":[{"id":"https://openalex.org/I4210086143","display_name":"Alibaba Group (Cayman Islands)","ror":"https://ror.org/00mnrxf72","country_code":"KY","type":"company","lineage":["https://openalex.org/I4210086143","https://openalex.org/I45928872"]}],"countries":["KY"],"is_corresponding":false,"raw_author_name":"Wotao Yin","raw_affiliation_strings":["DAMO Academy , Alibaba group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"DAMO Academy , Alibaba group","institution_ids":["https://openalex.org/I4210086143"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"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":0.29161522,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"7806","last_page":"7817"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.6338000297546387,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.6338000297546387,"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/T12101","display_name":"Advanced Bandit Algorithms Research","score":0.030500000342726707,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T12072","display_name":"Machine Learning and Algorithms","score":0.019099999219179153,"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":0.559499979019165},{"id":"https://openalex.org/keywords/scale","display_name":"Scale (ratio)","score":0.5092999935150146},{"id":"https://openalex.org/keywords/selection","display_name":"Selection (genetic algorithm)","score":0.49050000309944153},{"id":"https://openalex.org/keywords/stability","display_name":"Stability (learning theory)","score":0.32899999618530273},{"id":"https://openalex.org/keywords/control","display_name":"Control (management)","score":0.32179999351501465},{"id":"https://openalex.org/keywords/feature-selection","display_name":"Feature selection","score":0.3066999912261963},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.2955000102519989}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6211000084877014},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.559499979019165},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.5092999935150146},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.49050000309944153},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4717999994754791},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.39419999718666077},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.32899999618530273},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.32179999351501465},{"id":"https://openalex.org/C148483581","wikidata":"https://www.wikidata.org/wiki/Q446488","display_name":"Feature selection","level":2,"score":0.3066999912261963},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.2955000102519989},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2903999984264374},{"id":"https://openalex.org/C163985040","wikidata":"https://www.wikidata.org/wiki/Q1172399","display_name":"Data acquisition","level":2,"score":0.27090001106262207},{"id":"https://openalex.org/C196340769","wikidata":"https://www.wikidata.org/wiki/Q7698910","display_name":"Temporal difference learning","level":3,"score":0.27090001106262207},{"id":"https://openalex.org/C197298091","wikidata":"https://www.wikidata.org/wiki/Q5318963","display_name":"Dynamic data","level":2,"score":0.26750001311302185},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.2581000030040741},{"id":"https://openalex.org/C166109690","wikidata":"https://www.wikidata.org/wiki/Q4677422","display_name":"Action selection","level":3,"score":0.2515999972820282},{"id":"https://openalex.org/C133462117","wikidata":"https://www.wikidata.org/wiki/Q4929239","display_name":"Data collection","level":2,"score":0.25110000371932983}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-emnlp.412","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.412","pdf_url":"https://aclanthology.org/2025.findings-emnlp.412.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: EMNLP 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.412","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.412","pdf_url":"https://aclanthology.org/2025.findings-emnlp.412.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: EMNLP 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320336324","display_name":"State Grid Shandong Electric Power Company","ror":null}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416034889.pdf","grobid_xml":"https://content.openalex.org/works/W4416034889.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Optimizing":[0],"data":[1,110,119],"utilization":[2],"remains":[3],"a":[4,47],"central":[5],"challenge":[6],"in":[7],"applying":[8],"Reinforcement":[9],"Learning":[10],"(RL)":[11],"to":[12,35,59,77,121,136,154],"Large":[13],"Language":[14],"Models":[15],"(LLMs),":[16],"directly":[17],"impacting":[18],"sample":[19],"efficiency,":[20],"training":[21,57,104,138,149],"stability,":[22],"and":[23,38,85,100,127],"final":[24],"model":[25],"performance.Current":[26],"approaches":[27],"often":[28],"rely":[29],"on":[30,158],"massive":[31],"static":[32],"datasets,":[33],"leading":[34],"computational":[36],"inefficiency":[37],"redundant":[39],"gradient":[40],"updates.In":[41],"this":[42],"paper,":[43],"we":[44],"propose":[45],"ScalingRL,":[46],"data-centric":[48],"RL":[49,61],"framework":[50],"that":[51,72,94,131],"dynamically":[52],"selects":[53],"the":[54,67,97,118,122],"most":[55],"informative":[56],"samples":[58,144],"optimize":[60],"for":[62],"mathematical":[63],"reasoning.Specifically,":[64],"Scal-ingRL":[65],"introduces":[66],"Data":[68],"Effectiveness":[69],"Score":[70],"(DES)":[71],"quantitatively":[73],"ranks":[74],"prompts":[75],"according":[76],"three":[78],"complementary":[79],"factors:":[80],"problem":[81],"difficulty,":[82],"Chain-of-Thought":[83],"complexity,":[84],"reward":[86],"adaptability.Then,":[87],"ScalingRL":[88,132],"employs":[89],"an":[90],"adaptive":[91],"curriculum":[92],"scheduler":[93],"progressively":[95],"adjusts":[96],"overall":[98],"scale":[99],"specific":[101],"mix":[102],"of":[103,107,113,146],"prompts-balancing":[105],"exploration":[106],"new,":[108],"challenging":[109],"with":[111],"exploitation":[112],"previously":[114],"learned":[115],"concepts-thereby":[116],"tailoring":[117],"distribution":[120],"model's":[123],"current":[124],"learning":[125],"trajectory":[126],"performance.Experimental":[128],"results":[129],"demonstrate":[130],"achieves":[133],"comparable":[134],"performance":[135],"full-data":[137],"methods":[139],"while":[140],"requiring":[141],"only":[142],"1.5K":[143],"instead":[145],"220K,":[147],"reducing":[148],"time":[150],"from":[151],"13":[152],"days":[153],"just":[155],"4":[156],"hours":[157],"8\u00d7A800":[159],"GPUs.":[160]},"counts_by_year":[],"updated_date":"2026-08-21T09:56:20.448147","created_date":"2025-11-08T00:00:00"}