{"id":"https://openalex.org/W7221892412","doi":"https://doi.org/10.48550/arxiv.2610.19176","title":"Learning from Self-Generated Responses: Reliability--Uncertainty Self-Distillation for Reasoning","display_name":"Learning from Self-Generated Responses: Reliability--Uncertainty Self-Distillation for Reasoning","publication_year":2026,"publication_date":"2026-10-13","ids":{"openalex":"https://openalex.org/W7221892412","doi":"https://doi.org/10.48550/arxiv.2610.19176"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.19176","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.19176","pdf_url":"https://arxiv.org/pdf/2610.19176","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-nc","license_id":"https://openalex.org/licenses/cc-by-nc","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/2610.19176","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5158779083","display_name":"Jinhao Pan","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pan, Jinhao","raw_affiliation_strings":["University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129711177","display_name":"Bowen Wei","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wei, Bowen","raw_affiliation_strings":["University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100951608","display_name":"Yuqing Zhou","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Yuqing","raw_affiliation_strings":["University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5158831619","display_name":"Ziwei Zhu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhu, Ziwei","raw_affiliation_strings":["University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University","institution_ids":[]}]}],"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":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":0.8579000234603882,"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":0.8579000234603882,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.035999998450279236,"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":0.020600000396370888,"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/self-distillation","display_name":"self-distillation","score":0.9440000057220459},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.6859999895095825},{"id":"https://openalex.org/keywords/reasoning","display_name":"reasoning","score":0.39399999380111694},{"id":"https://openalex.org/keywords/reliability-estimation","display_name":"reliability estimation","score":0.5389999747276306},{"id":"https://openalex.org/keywords/forecast-uncertainty","display_name":"forecast uncertainty","score":0.8050000071525574},{"id":"https://openalex.org/keywords/teacher-student-learning","display_name":"teacher-student learning","score":0.7440000176429749},{"id":"https://openalex.org/keywords/open-ended-tasks","display_name":"open-ended tasks","score":0.6740000247955322},{"id":"https://openalex.org/keywords/knowledge-distillation","display_name":"knowledge distillation","score":0.5379999876022339}],"concepts":[{"id":"https://openalex.org/C111640148","wikidata":"https://www.wikidata.org/wiki/Q847349","display_name":"Rubric","level":2,"score":0.8597000241279602},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6929000020027161},{"id":"https://openalex.org/C55439883","wikidata":"https://www.wikidata.org/wiki/Q360812","display_name":"Correctness","level":2,"score":0.6658999919891357},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5490999817848206},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.446399986743927},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4366999864578247},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.4336000084877014},{"id":"https://openalex.org/C42058472","wikidata":"https://www.wikidata.org/wiki/Q810214","display_name":"Base (topology)","level":2,"score":0.4000000059604645},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.39969998598098755},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.3856000006198883},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.35510000586509705},{"id":"https://openalex.org/C45235069","wikidata":"https://www.wikidata.org/wiki/Q278425","display_name":"Table (database)","level":2,"score":0.2994999885559082},{"id":"https://openalex.org/C4554734","wikidata":"https://www.wikidata.org/wiki/Q593744","display_name":"Knowledge base","level":2,"score":0.28619998693466187},{"id":"https://openalex.org/C20162079","wikidata":"https://www.wikidata.org/wiki/Q1151406","display_name":"Case-based reasoning","level":2,"score":0.2612999975681305},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.25839999318122864}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2610.19176","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.19176","pdf_url":"https://arxiv.org/pdf/2610.19176","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-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2610.19176","doi":"https://doi.org/10.48550/arxiv.2610.19176","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2610.19176","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:2610.19176","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.19176","pdf_url":"https://arxiv.org/pdf/2610.19176","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-nc","license_id":"https://openalex.org/licenses/cc-by-nc","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":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7221892412.pdf","grobid_xml":"https://content.openalex.org/works/W7221892412.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7162531820","https://openalex.org/W7222515981","https://openalex.org/W7162093714","https://openalex.org/W7158506380","https://openalex.org/W7125861034","https://openalex.org/W7128037996","https://openalex.org/W7202201236","https://openalex.org/W7160995547","https://openalex.org/W4415967455","https://openalex.org/W4415272049"],"abstract_inverted_index":{"Reference":[0],"solutions":[1],"and":[2,39,64,71,74,102,124,129,138],"question-specific":[3],"rubrics":[4],"are":[5,140],"not":[6],"always":[7],"available":[8,141],"to":[9,30,53],"guide":[10],"the":[11,66,69,75,78,134],"teacher":[12,63,67],"in":[13,32,117],"self-distillation.":[14],"A":[15],"large":[16],"language":[17],"model's":[18],"own":[19],"solution":[20],"attempts":[21,42],"offer":[22],"an":[23],"alternative,":[24],"but":[25],"their":[26],"quality":[27],"is":[28],"difficult":[29],"assess":[31],"open-ended":[33],"tasks":[34],"without":[35],"automatic":[36],"correctness":[37],"checks,":[38],"even":[40],"correct":[41],"can":[43],"involve":[44],"uncertain":[45],"token":[46],"predictions.":[47],"We":[48],"propose":[49],"Reliability--Uncertainty":[50],"Self-Distillation":[51],"(\\method{})":[52],"learn":[54],"from":[55,98],"these":[56],"attempts.":[57],"The":[58],"same":[59],"model":[60],"serves":[61],"as":[62],"student:":[65],"receives":[68],"question":[70],"completed":[72],"attempt,":[73],"student":[76,105],"matches":[77],"teacher's":[79],"token-level":[80],"predictions":[81],"using":[82],"a":[83,99],"question-only":[84],"prompt.":[85],"\\method{}":[86,112],"weights":[87],"each":[88],"attempt's":[89],"distillation":[90],"loss":[91],"by":[92],"additively":[93],"combining":[94],"judged":[95],"reliability,":[96],"estimated":[97],"self-judge's":[100],"score":[101],"confidence,":[103],"with":[104],"predictive":[106],"uncertainty.":[107],"Across":[108],"eleven":[109],"reasoning":[110],"benchmarks,":[111],"outperforms":[113],"all":[114],"five":[115],"baselines":[116],"average":[118],"accuracy,":[119],"reaching":[120],"70.47\\%":[121],"on":[122,126],"Qwen3-4B":[123],"76.69\\%":[125],"Qwen3-8B":[127],"(8.90":[128],"2.10":[130],"percentage":[131],"points":[132],"above":[133],"base":[135],"models).":[136],"Data":[137],"code":[139],"at":[142],"https://github.com/JP-25/RUSD.":[143]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-10-16T00:00:00"}