{"id":"https://openalex.org/W7204752623","doi":"https://doi.org/10.48550/arxiv.2608.28306","title":"VISTA: Verifier-Informed Student-to-Teacher Adaptation for On-Policy Self-Distillation","display_name":"VISTA: Verifier-Informed Student-to-Teacher Adaptation for On-Policy Self-Distillation","publication_year":2026,"publication_date":"2026-08-28","ids":{"openalex":"https://openalex.org/W7204752623","doi":"https://doi.org/10.48550/arxiv.2608.28306"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2608.28306","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.28306","pdf_url":"https://arxiv.org/pdf/2608.28306","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,"listed_in":[],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2608.28306","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5122401378","display_name":"Zewen Ding","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ding, Zewen","raw_affiliation_strings":["State Key Laboratory of Cognitive Intelligence","University of Science and Technology of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of Cognitive Intelligence","institution_ids":[]},{"raw_affiliation_string":"University of Science and Technology of China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5122383397","display_name":"Zezhong Wu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wu, Zezhong","raw_affiliation_strings":["State Key Laboratory of Cognitive Intelligence","University of Science and Technology of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of Cognitive Intelligence","institution_ids":[]},{"raw_affiliation_string":"University of Science and Technology of China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130954508","display_name":"Zhou Tao","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tao, Zhou","raw_affiliation_strings":["State Key Laboratory of Cognitive Intelligence","University of Science and Technology of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of Cognitive Intelligence","institution_ids":[]},{"raw_affiliation_string":"University of Science and Technology of China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5148808760","display_name":"Shida Wang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wang, Shida","raw_affiliation_strings":["State Key Laboratory of Cognitive Intelligence","University of Science and Technology of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of Cognitive Intelligence","institution_ids":[]},{"raw_affiliation_string":"University of Science and Technology of China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5148770351","display_name":"Shizhuo Hou","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hou, Shizhuo","raw_affiliation_strings":["State Key Laboratory of Cognitive Intelligence","University of Science and Technology of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of Cognitive Intelligence","institution_ids":[]},{"raw_affiliation_string":"University of Science and Technology of China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042417337","display_name":"Yupeng Hua","orcid":"https://orcid.org/0009-0008-8849-4717","observed_orcids":["https://orcid.org/0009-0008-8849-4717"]},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hua, YongXiang","raw_affiliation_strings":["State Key Laboratory of Cognitive Intelligence","University of Science and Technology of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of Cognitive Intelligence","institution_ids":[]},{"raw_affiliation_string":"University of Science and Technology of China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129999900","display_name":"Haoyu Cao","orcid":"https://orcid.org/0000-0002-3789-9705","observed_orcids":["https://orcid.org/0000-0002-3789-9705"]},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Cao, Haoyu","raw_affiliation_strings":["State Key Laboratory of Cognitive Intelligence","University of Science and Technology of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of Cognitive Intelligence","institution_ids":[]},{"raw_affiliation_string":"University of Science and Technology of China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101859765","display_name":"Linli Xu","orcid":"https://orcid.org/0000-0001-6004-4425","observed_orcids":["https://orcid.org/0000-0001-6004-4425"]},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xu, Linli","raw_affiliation_strings":["State Key Laboratory of Cognitive Intelligence","University of Science and Technology of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of Cognitive Intelligence","institution_ids":[]},{"raw_affiliation_string":"University of Science and Technology of China","institution_ids":["https://openalex.org/I126520041"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I126520041"],"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":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.3296000063419342,"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/T10028","display_name":"Topic Modeling","score":0.3296000063419342,"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/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.24390000104904175,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.05290000140666962,"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/on-policy-self-distillation","display_name":"on-policy self-distillation","score":0.9980000257492065},{"id":"https://openalex.org/keywords/kullback-leibler-divergence","display_name":"Kullback-Leibler divergence","score":0.6629999876022339},{"id":"https://openalex.org/keywords/large-language-model-reasoning","display_name":"large language model reasoning","score":0.5490000247955322},{"id":"https://openalex.org/keywords/aime","display_name":"AIME","score":0.4749999940395355}],"concepts":[{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.6193000078201294},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.5924999713897705},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5379999876022339},{"id":"https://openalex.org/C110121322","wikidata":"https://www.wikidata.org/wiki/Q865811","display_name":"Distribution (mathematics)","level":2,"score":0.3797000050544739},{"id":"https://openalex.org/C2776291640","wikidata":"https://www.wikidata.org/wiki/Q2912517","display_name":"Value (mathematics)","level":2,"score":0.3797000050544739},{"id":"https://openalex.org/C3020207766","wikidata":"https://www.wikidata.org/wiki/Q4405230","display_name":"Student teacher","level":3,"score":0.3628000020980835},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.3273000121116638},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.3183000087738037},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.2957000136375427},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.29170000553131104},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.28439998626708984},{"id":"https://openalex.org/C9652623","wikidata":"https://www.wikidata.org/wiki/Q190109","display_name":"Field (mathematics)","level":2,"score":0.2754000127315521},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.2754000127315521},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.26930001378059387},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.26759999990463257},{"id":"https://openalex.org/C2776905826","wikidata":"https://www.wikidata.org/wiki/Q22909282","display_name":"Teacher quality","level":3,"score":0.2524000108242035}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.28306","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.28306","pdf_url":"https://arxiv.org/pdf/2608.28306","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,"listed_in":[],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2608.28306","doi":"https://doi.org/10.48550/arxiv.2608.28306","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.28306","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,"listed_in":[],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2608.28306","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.28306","pdf_url":"https://arxiv.org/pdf/2608.28306","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,"listed_in":[],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","display_name":"Quality education","score":0.8278999924659729}],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7204752623.pdf","grobid_xml":"https://content.openalex.org/works/W7204752623.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7202213474","https://openalex.org/W7214731163","https://openalex.org/W7215090909","https://openalex.org/W7223337476","https://openalex.org/W7204449334","https://openalex.org/W7220652251","https://openalex.org/W7220699741","https://openalex.org/W7220388840","https://openalex.org/W7222213547","https://openalex.org/W7172413156"],"abstract_inverted_index":{"On-policy":[0],"self-distillation":[1],"(OPSD)":[2],"improves":[3],"reasoning":[4],"by":[5,176],"training":[6],"a":[7,19,25,36,199],"problem-only":[8,68],"student":[9,47,77,86,100,112,189],"on":[10],"its":[11],"own":[12],"rollouts":[13,105,193],"using":[14,103],"dense":[15],"token-level":[16],"supervision":[17,72,190],"from":[18,141,191],"privileged":[20,53],"teacher":[21,33,60,80,109],"that":[22,58],"also":[23],"sees":[24],"reference":[26],"solution.":[27],"However,":[28],"standard":[29,98,142],"OPSD":[30,99,175],"treats":[31],"the":[32,40,46,59,63,76,79,97,108,111,124,128,136,167,186],"distribution":[34,81],"as":[35,198],"fixed":[37],"target":[38,66],"along":[39],"student's":[41],"rollout":[42,137],"and":[43,138,155,163,179,194],"updates":[44],"only":[45],"%,":[48],"although":[49],"--":[50],"even":[51],"though":[52],"conditioning":[54],"does":[55],"not":[56],"guarantee":[57],"always":[61],"provides":[62],"most":[64],"appropriate":[65],"for":[67,202],"reasoning.":[69,87],"This":[70],"one-way":[71],"can":[73],"therefore":[74,89],"misdirect":[75],"when":[78],"is":[82],"misaligned":[83],"with":[84,127,157],"valid":[85],"We":[88],"introduce":[90],"Verifier-Informed":[91],"Student-to-Teacher":[92],"Adaptation":[93],"(VISTA),":[94],"which":[95],"preserves":[96],"update":[101],"while":[102],"outcome-verified":[104,192],"to":[106,123],"adapt":[107],"toward":[110],"distribution.":[113],"Within":[114],"each":[115],"verified":[116],"rollout,":[117],"VISTA":[118,134,165],"further":[119],"restricts":[120],"this":[121],"adaptation":[122,197],"top-$k$":[125],"positions":[126],"largest":[129],"teacher--student":[130],"KL":[131],"divergence.":[132],"Notably,":[133],"reuses":[135],"loss":[139],"function":[140],"OPSD,":[143],"introducing":[144],"no":[145],"additional":[146],"sampling":[147],"or":[148],"separate":[149],"reward":[150],"objective.":[151],"Across":[152],"AIME24,":[153],"AIME25,":[154],"HMMT25":[156],"Qwen3":[158],"models":[159],"at":[160,170],"1.7B,":[161],"4B,":[162],"8B,":[164],"achieves":[166],"highest":[168],"Avg@12":[169],"every":[171],"scale,":[172],"improving":[173],"over":[174],"$0.6$,":[177],"$0.7$,":[178],"$2.1$":[180],"points,":[181],"respectively.":[182],"These":[183],"results":[184],"demonstrate":[185],"value":[187],"of":[188],"highlight":[195],"student-to-teacher":[196],"promising":[200],"direction":[201],"OPSD.":[203]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-09-01T00:00:00"}
