{"id":"https://openalex.org/W7225834324","doi":"https://doi.org/10.48550/arxiv.2611.14515","title":"AutoScaffold: AI-Guided Scaffolding for Reinforcement Learning of Language Models","display_name":"AutoScaffold: AI-Guided Scaffolding for Reinforcement Learning of Language Models","publication_year":2026,"publication_date":"2026-11-10","ids":{"openalex":"https://openalex.org/W7225834324","doi":"https://doi.org/10.48550/arxiv.2611.14515"},"language":"en","primary_location":{"id":"doi:10.48550/arxiv.2611.14515","doi":"https://doi.org/10.48550/arxiv.2611.14515","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2611.14515","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2611.14515","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5161458212","display_name":"Zijian Zhang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Zijian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5121322693","display_name":"Rizhen Hu","orcid":"https://orcid.org/0009-0000-3812-1252","observed_orcids":["https://orcid.org/0009-0000-3812-1252"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hu, Rizhen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126391805","display_name":"Jiangweizhi Peng","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Peng, Jiangweizhi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5161464103","display_name":"Shiyang Li","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Shiyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5083631658","display_name":"Sujay Bhatt","orcid":"https://orcid.org/0000-0002-3586-1789","observed_orcids":["https://orcid.org/0000-0002-3586-1789"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bhatt, Sujay","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5069872754","display_name":"Sihan Zeng","orcid":"https://orcid.org/0000-0002-0061-5780","observed_orcids":["https://orcid.org/0000-0002-0061-5780"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zeng, Sihan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5100633783","display_name":"Mingyi Hong","orcid":"https://orcid.org/0000-0003-1263-9365","observed_orcids":["https://orcid.org/0000-0003-1263-9365"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hong, Mingyi","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":null,"has_fulltext":false,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.923799991607666,"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":0.923799991607666,"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":0.016100000590085983,"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":0.015699999406933784,"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":[],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7386000156402588},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6998999714851379},{"id":"https://openalex.org/C177212765","wikidata":"https://www.wikidata.org/wiki/Q627335","display_name":"Workflow","level":2,"score":0.6092000007629395},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.5785999894142151},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.5360000133514404},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.5295000076293945},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.49410000443458557},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.4530999958515167},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4447000026702881},{"id":"https://openalex.org/C149629883","wikidata":"https://www.wikidata.org/wiki/Q660926","display_name":"Fraction (chemistry)","level":2,"score":0.40540000796318054},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.3984000086784363},{"id":"https://openalex.org/C148220186","wikidata":"https://www.wikidata.org/wiki/Q7111912","display_name":"Outcome (game theory)","level":2,"score":0.3889000117778778},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.3871999979019165},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.3732999861240387},{"id":"https://openalex.org/C2994555788","wikidata":"https://www.wikidata.org/wiki/Q5689226","display_name":"Head start","level":2,"score":0.37130001187324524},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.31790000200271606},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.31619998812675476},{"id":"https://openalex.org/C77967617","wikidata":"https://www.wikidata.org/wiki/Q4677561","display_name":"Active learning (machine learning)","level":2,"score":0.3151000142097473},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.29170000553131104},{"id":"https://openalex.org/C60048249","wikidata":"https://www.wikidata.org/wiki/Q37437","display_name":"Syntax","level":2,"score":0.2888000011444092},{"id":"https://openalex.org/C85847156","wikidata":"https://www.wikidata.org/wiki/Q59015987","display_name":"Verifiable secret sharing","level":3,"score":0.275299996137619},{"id":"https://openalex.org/C29122968","wikidata":"https://www.wikidata.org/wiki/Q1414816","display_name":"Incentive","level":2,"score":0.27300000190734863},{"id":"https://openalex.org/C206345919","wikidata":"https://www.wikidata.org/wiki/Q20380951","display_name":"Resource (disambiguation)","level":2,"score":0.26260000467300415},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.2540000081062317},{"id":"https://openalex.org/C2778583658","wikidata":"https://www.wikidata.org/wiki/Q849415","display_name":"On-the-job training","level":3,"score":0.2524999976158142}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2611.14515","doi":"https://doi.org/10.48550/arxiv.2611.14515","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2611.14515","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2611.14515","doi":"https://doi.org/10.48550/arxiv.2611.14515","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2611.14515","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Reinforcement":[0],"learning":[1],"with":[2,81,159,164,173,199],"verifiable":[3],"rewards":[4,12],"(RLVR)":[5],"produces":[6],"rich":[7],"training":[8,76,229],"experience,":[9,97],"yet":[10],"outcome":[11],"reduce":[13],"each":[14,241],"attempt":[15],"to":[16,35,44,58,62,143,212,251],"a":[17,41,68,74,110,200,237],"single":[18],"number":[19],"and":[20,53,60,85,100,114,124,149,181,244,258],"leave":[21],"most":[22],"of":[23,112,126,175,197],"this":[24,46,96],"information":[25,47,84],"unused.":[26],"Training-time":[27],"scaffolds,":[28,80,156],"such":[29],"as":[30,128,226,254],"strategies":[31,107],"or":[32,273],"hints":[33,116],"added":[34],"the":[36,122,129,140,165,195,207,234,255,264,270],"prompt":[37],"during":[38,89],"training,":[39],"offer":[40],"natural":[42],"way":[43],"turn":[45],"into":[48,78],"guidance,":[49],"but":[50],"their":[51],"design":[52],"adaptation":[54],"are":[55,134],"largely":[56],"left":[57],"humans":[59],"tailored":[61],"specific":[63],"domains.":[64],"We":[65,138],"introduce":[66],"AutoScaffold,":[67],"unified":[69],"AI-for-AI":[70],"framework":[71,142],"that":[72,203,233,259],"turns":[73],"student's":[75],"experience":[77],"adaptive":[79],"minimal":[82],"domain-specific":[83],"no":[86,275],"human":[87],"intervention":[88],"training.":[90],"A":[91],"frozen":[92],"teacher":[93,235],"periodically":[94],"examines":[95],"diagnoses":[98],"bottlenecks,":[99],"designs":[101],"scaffolds":[102,133,247,262],"in":[103,169,178,184,240],"two":[104],"forms:":[105],"(i)":[106],"shared":[108],"across":[109],"class":[111,243],"tasks":[113,198],"(ii)":[115],"for":[117],"individual":[118],"hard":[119],"tasks,":[120],"adjusting":[121],"content":[123],"amount":[125],"support":[127],"student":[130,218,228,256,265],"learns.":[131],"The":[132],"removed":[135],"at":[136],"evaluation.":[137],"apply":[139],"same":[141],"ALFWorld,":[144],"search-based":[145],"QA,":[146],"mathematical":[147],"reasoning,":[148],"GPU":[150,190],"kernel":[151,191,202],"generation.":[152],"When":[153],"evaluated":[154],"without":[155],"students":[157],"trained":[158,163],"AutoScaffold":[160,193],"outperform":[161],"those":[162],"corresponding":[166],"RL":[167],"baselines":[168],"all":[170],"four":[171],"domains,":[172],"gains":[174],"7.2":[176],"points":[177,183],"ALFWorld":[179],"success":[180],"4.0":[182],"average":[185],"QA":[186],"exact":[187],"match.":[188],"On":[189,215],"generation,":[192],"increases":[194],"fraction":[196],"correct":[201],"runs":[204],"faster":[205],"than":[206,268],"PyTorch":[208],"reference":[209],"by":[210],"up":[211],"5.0":[213],"points.":[214],"math,":[216],"our":[217],"surpasses":[219],"QuestA's":[220],"reported":[221],"accuracy":[222],"using":[223,274],"approximately":[224],"one-third":[225],"many":[227],"steps.":[230],"Analyses":[231],"show":[232],"identifies":[236],"different":[238],"failure":[239],"task":[242],"shifts":[245],"its":[246,260],"from":[248],"step-by-step":[249],"workflows":[250],"targeted":[252],"fixes":[253],"improves,":[257],"revised":[261],"help":[263],"learn":[266],"more":[267],"keeping":[269],"previous":[271],"scaffold":[272],"scaffold.":[276]},"counts_by_year":[],"updated_date":"2026-11-12T12:28:57.059296","created_date":"2026-11-12T00:00:00"}