{"id":"https://openalex.org/W7138915204","doi":"https://doi.org/10.1609/aaai.v40i40.40710","title":"RLKD: Distilling LLMs\u2019 Reasoning via Reinforcement Learning","display_name":"RLKD: Distilling LLMs\u2019 Reasoning via Reinforcement Learning","publication_year":2026,"publication_date":"2026-03-14","ids":{"openalex":"https://openalex.org/W7138915204","doi":"https://doi.org/10.1609/aaai.v40i40.40710"},"language":null,"primary_location":{"id":"doi:10.1609/aaai.v40i40.40710","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i40.40710","pdf_url":"https://ojs.aaai.org/index.php/AAAI/article/download/40710/44671","source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://ojs.aaai.org/index.php/AAAI/article/download/40710/44671","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Shicheng Xu","orcid":null},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shicheng Xu","raw_affiliation_strings":["State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences\nUniversity of Chinese Academy of Sciences"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences\nUniversity of Chinese Academy of Sciences","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Liang Pang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Liang Pang","raw_affiliation_strings":["State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Yunchang Zhu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yunchang Zhu","raw_affiliation_strings":["Huawei Inc"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huawei Inc","institution_ids":[]}]},{"author_position":"middle","author":{"id":null,"display_name":"Jia Gu","orcid":null},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jia Gu","raw_affiliation_strings":["State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences\nUniversity of Chinese Academy of Sciences"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences\nUniversity of Chinese Academy of Sciences","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Zihao Wei","orcid":null},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zihao Wei","raw_affiliation_strings":["State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences\nUniversity of Chinese Academy of Sciences"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences\nUniversity of Chinese Academy of Sciences","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Jingcheng Deng","orcid":null},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jingcheng Deng","raw_affiliation_strings":["State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences\nUniversity of Chinese Academy of Sciences"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences\nUniversity of Chinese Academy of Sciences","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Feiyang Pan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Feiyang Pan","raw_affiliation_strings":["Huawei Inc"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huawei Inc","institution_ids":[]}]},{"author_position":"middle","author":{"id":null,"display_name":"Huawei Shen","orcid":null},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Huawei Shen","raw_affiliation_strings":["State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"last","author":{"id":null,"display_name":"Xueqi Cheng","orcid":null},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xueqi Cheng","raw_affiliation_strings":["State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences","institution_ids":["https://openalex.org/I4210090176"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"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":"40","issue":"40","first_page":"34151","last_page":"34159"},"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":0.31290000677108765,"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":0.31290000677108765,"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/T10028","display_name":"Topic Modeling","score":0.15469999611377716,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.08240000158548355,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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":0.5726000070571899},{"id":"https://openalex.org/keywords/generative-grammar","display_name":"Generative grammar","score":0.47920000553131104},{"id":"https://openalex.org/keywords/qualitative-reasoning","display_name":"Qualitative reasoning","score":0.4424000084400177},{"id":"https://openalex.org/keywords/commonsense-reasoning","display_name":"Commonsense reasoning","score":0.4372999966144562},{"id":"https://openalex.org/keywords/opportunistic-reasoning","display_name":"Opportunistic reasoning","score":0.4194999933242798},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.4115000069141388},{"id":"https://openalex.org/keywords/sequence","display_name":"Sequence (biology)","score":0.39329999685287476},{"id":"https://openalex.org/keywords/cognition","display_name":"Cognition","score":0.3910999894142151},{"id":"https://openalex.org/keywords/spatial-intelligence","display_name":"Spatial intelligence","score":0.39070001244544983}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.663100004196167},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5781999826431274},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.5726000070571899},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.47920000553131104},{"id":"https://openalex.org/C83725634","wikidata":"https://www.wikidata.org/wiki/Q7268699","display_name":"Qualitative reasoning","level":2,"score":0.4424000084400177},{"id":"https://openalex.org/C193221554","wikidata":"https://www.wikidata.org/wiki/Q5153664","display_name":"Commonsense reasoning","level":2,"score":0.4372999966144562},{"id":"https://openalex.org/C86827895","wikidata":"https://www.wikidata.org/wiki/Q7098582","display_name":"Opportunistic reasoning","level":4,"score":0.4194999933242798},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.4115000069141388},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.39329999685287476},{"id":"https://openalex.org/C169900460","wikidata":"https://www.wikidata.org/wiki/Q2200417","display_name":"Cognition","level":2,"score":0.3910999894142151},{"id":"https://openalex.org/C155911833","wikidata":"https://www.wikidata.org/wiki/Q3817354","display_name":"Spatial intelligence","level":2,"score":0.39070001244544983},{"id":"https://openalex.org/C89288958","wikidata":"https://www.wikidata.org/wiki/Q7301504","display_name":"Reasoning system","level":2,"score":0.3711000084877014},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.36649999022483826},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.3474000096321106},{"id":"https://openalex.org/C2777735758","wikidata":"https://www.wikidata.org/wiki/Q817765","display_name":"Path (computing)","level":2,"score":0.31679999828338623},{"id":"https://openalex.org/C115086926","wikidata":"https://www.wikidata.org/wiki/Q17004651","display_name":"Causal reasoning","level":3,"score":0.3142000138759613},{"id":"https://openalex.org/C36964233","wikidata":"https://www.wikidata.org/wiki/Q7920942","display_name":"Verbal reasoning","level":3,"score":0.3043000102043152},{"id":"https://openalex.org/C183521366","wikidata":"https://www.wikidata.org/wiki/Q7256422","display_name":"Psychology of reasoning","level":4,"score":0.30379998683929443},{"id":"https://openalex.org/C188147891","wikidata":"https://www.wikidata.org/wiki/Q147638","display_name":"Cognitive science","level":1,"score":0.3019999861717224},{"id":"https://openalex.org/C103057564","wikidata":"https://www.wikidata.org/wiki/Q4751139","display_name":"Analytic reasoning","level":3,"score":0.29899999499320984},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2930999994277954},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.2888999879360199},{"id":"https://openalex.org/C159032336","wikidata":"https://www.wikidata.org/wiki/Q2488768","display_name":"Non-monotonic logic","level":2,"score":0.28839999437332153},{"id":"https://openalex.org/C107848011","wikidata":"https://www.wikidata.org/wiki/Q4680756","display_name":"Adaptive reasoning","level":4,"score":0.2678000032901764},{"id":"https://openalex.org/C37335422","wikidata":"https://www.wikidata.org/wiki/Q6888134","display_name":"Model-based reasoning","level":3,"score":0.26249998807907104},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2572000026702881},{"id":"https://openalex.org/C195344581","wikidata":"https://www.wikidata.org/wiki/Q2555318","display_name":"Automated reasoning","level":2,"score":0.2556999921798706}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1609/aaai.v40i40.40710","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i40.40710","pdf_url":"https://ojs.aaai.org/index.php/AAAI/article/download/40710/44671","source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1609/aaai.v40i40.40710","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i40.40710","pdf_url":"https://ojs.aaai.org/index.php/AAAI/article/download/40710/44671","source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","score":0.7992057800292969,"display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7138915204.pdf","grobid_xml":"https://content.openalex.org/works/W7138915204.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Distilling":[0],"reasoning":[1,18,29,52,77,97,132,148,212],"paths":[2,30],"from":[3,46,64],"teacher":[4,33],"to":[5,104,142,165],"student":[6,151,163,216],"models":[7,34],"via":[8],"supervised":[9],"fine-tuning":[10,83],"(SFT)":[11],"provides":[12],"a":[13,54,89,113,121],"shortcut":[14],"for":[15],"improving":[16],"the":[17,21,28,61,71,131,140,144,147,162,167,194,201,210,215],"ability":[19,213],"of":[20,40,92,150,193,203,214],"smaller":[22],"Large":[23],"Language":[24],"Models":[25],"(LLMs).":[26],"However,":[27],"generated":[31],"by":[32,120],"often":[35],"reflect":[36],"only":[37,191],"surface-level":[38],"traces":[39],"their":[41],"underlying":[42],"authentic":[43,51,76,173],"reasoning.":[44],"Insights":[45],"cognitive":[47],"neuroscience":[48],"suggest":[49],"that":[50,59,75,185],"involves":[53],"complex":[55],"interweaving":[56],"between":[57,146],"meta-reasoning":[58],"selects":[60],"appropriate":[62],"sub-problem":[63],"multiple":[65,135],"candidates,":[66],"and":[67,138,152,207],"solving,":[68],"which":[69,99],"addresses":[70],"sub-problem.":[72],"It":[73],"means":[74],"has":[78],"implicit":[79,169],"multi-branch":[80,170],"structure.":[81],"Supervised":[82],"collapses":[84],"this":[85,102,108,157],"rich":[86],"structure":[87,103,171],"into":[88,134],"flat":[90],"sequence":[91],"token":[93],"prediction":[94],"in":[95,172],"teacher's":[96,180],"path,":[98],"cannot":[100],"distill":[101],"student.":[105],"To":[106],"address":[107],"limitation,":[109],"we":[110],"propose":[111],"RLKD,":[112,186],"reinforcement":[114],"learning":[115],"(RL)-based":[116],"distillation":[117],"framework":[118],"guided":[119],"novel":[122],"Generative":[123],"Structure":[124],"Reward":[125],"Model":[126],"(GSRM).":[127],"Our":[128,154],"GSRM":[129],"converts":[130],"path":[133],"meta-reasoning-solving":[136],"steps":[137],"gives":[139],"reward":[141,158],"measure":[143],"alignment":[145],"structures":[149],"teacher.":[153],"RLKD":[155],"combines":[156],"with":[159],"RL,":[160],"enables":[161],"LLM":[164,217],"internalize":[166],"teacher\u2019s":[168],"reasoning,":[174],"rather":[175],"than":[176,218],"merely":[177],"mimicking":[178],"fixed":[179],"output":[181],"paths.":[182],"Experiments":[183],"show":[184],"even":[187],"when":[188],"trained":[189],"on":[190],"0.1%":[192],"data":[195],"under":[196],"an":[197],"RL-only":[198],"regime,":[199],"surpasses":[200],"performance":[202],"standard":[204],"SFT-RL":[205],"pipelines":[206],"further":[208],"unleashes":[209],"potential":[211],"SFT-based":[219],"distillation.":[220]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2026-02-06T00:00:00"}