{"id":"https://openalex.org/W7164367257","doi":"https://doi.org/10.48550/arxiv.2606.12373","title":"Verifiable Environments Are LEGO Bricks: Recursive Composition for Reasoning Generalization","display_name":"Verifiable Environments Are LEGO Bricks: Recursive Composition for Reasoning Generalization","publication_year":2026,"publication_date":"2026-06-10","ids":{"openalex":"https://openalex.org/W7164367257","doi":"https://doi.org/10.48550/arxiv.2606.12373"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.12373","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.12373","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.12373","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5138432083","display_name":"Hao Xiang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiang, Hao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138447104","display_name":"Qiaoyu Tang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tang, Qiaoyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138425530","display_name":"Le Yu (227543)","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Le","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138449354","display_name":"Yaojie Lu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lu, Yaojie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138398506","display_name":"Xianpei Han","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Han, Xianpei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138445025","display_name":"Ben He","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"He, Ben","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138411667","display_name":"Le Sun","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sun, Le","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138448531","display_name":"Bowen Yu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Bowen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138462084","display_name":"Peng Wang (34436)","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Peng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138464432","display_name":"Hongyu Lin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lin, Hongyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5138464646","display_name":"Dayiheng Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Dayiheng","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/T10028","display_name":"Topic Modeling","score":0.40950000286102295,"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":0.40950000286102295,"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":0.14350000023841858,"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"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.05889999866485596,"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/verifiable-secret-sharing","display_name":"Verifiable secret sharing","score":0.8946999907493591},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.5982000231742859},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.5913000106811523},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.5619000196456909},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.5116000175476074},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.49799999594688416},{"id":"https://openalex.org/keywords/base","display_name":"Base (topology)","score":0.40149998664855957},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.3862999975681305}],"concepts":[{"id":"https://openalex.org/C85847156","wikidata":"https://www.wikidata.org/wiki/Q59015987","display_name":"Verifiable secret sharing","level":3,"score":0.8946999907493591},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6978999972343445},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.5982000231742859},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.5913000106811523},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5619000196456909},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.5116000175476074},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.49799999594688416},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4702000021934509},{"id":"https://openalex.org/C42058472","wikidata":"https://www.wikidata.org/wiki/Q810214","display_name":"Base (topology)","level":2,"score":0.40149998664855957},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3944000005722046},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.3862999975681305},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.3643999993801117},{"id":"https://openalex.org/C40231798","wikidata":"https://www.wikidata.org/wiki/Q1333743","display_name":"Composition (language)","level":2,"score":0.3476000130176544},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.32429999113082886},{"id":"https://openalex.org/C2781020372","wikidata":"https://www.wikidata.org/wiki/Q533093","display_name":"On the fly","level":2,"score":0.3174999952316284},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2994999885559082},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.29789999127388},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.28850001096725464},{"id":"https://openalex.org/C4554734","wikidata":"https://www.wikidata.org/wiki/Q593744","display_name":"Knowledge base","level":2,"score":0.2685000002384186},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.2653000056743622}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.12373","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.12373","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.12373","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.12373","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Reinforcement":[0],"Learning":[1],"(RL)":[2],"with":[3,110],"verifiable":[4,63,102],"environments":[5,64,113,140,189],"has":[6],"emerged":[7],"as":[8,65],"a":[9,59,100,116],"powerful":[10],"approach":[11],"for":[12,56],"enhancing":[13],"the":[14,80,88,174],"reasoning":[15,47,129,143],"capabilities":[16],"of":[17,84,92,118,152,176],"Large":[18],"Language":[19],"Models":[20],"(LLMs).":[21],"While":[22],"prior":[23],"research":[24],"demonstrates":[25],"that":[26,61,69,78,126,134],"scaling":[27,42],"environment":[28,86,199],"quantity":[29],"improves":[30,147],"RL":[31,135],"performance,":[32],"existing":[33],"manual":[34],"or":[35],"individual":[36,112,188],"construction":[37,175],"methods":[38],"suffer":[39],"from":[40,163],"linear":[41],"limits,":[43],"thereby":[44],"hindering":[45],"scalable":[46],"generalization.":[48,144],"This":[49],"paper":[50],"introduces":[51],"RACES":[52,107,146,180],"(\\textbf{R}ecursive":[53],"\\textbf{A}utomated":[54],"\\textbf{C}omposition":[55],"\\textbf{E}nvironment":[57],"\\textbf{S}caling),":[58],"framework":[60],"conceptualizes":[62],"composable":[66],"building":[67],"blocks":[68],"can":[70,95],"be":[71,96],"recursively":[72],"assembled.":[73],"The":[74],"key":[75],"insight":[76],"is":[77,108],"when":[79],"codomain":[81],"(output":[82],"type)":[83,91],"one":[85],"matches":[87],"domain":[89],"(input":[90],"another,":[93],"they":[94],"automatically":[97],"fused":[98],"into":[99],"new":[101],"environment,":[103],"enabling":[104],"recursive":[105],"composition.":[106],"implemented":[109],"300":[111,187],"and":[114,124,159],"defines":[115],"set":[117],"composition":[119],"operators":[120],"(\\textsc{SEQUENTIAL},":[121],"\\textsc{PARALLEL},":[122],"\\textsc{SORT},":[123],"\\textsc{SELECT})":[125],"induce":[127],"diverse":[128],"patterns.":[130],"Extensive":[131],"experiments":[132],"show":[133],"training":[136,177,185],"on":[137,167,186],"these":[138],"composite":[139],"consistently":[141],"enhances":[142],"Specifically,":[145],"DeepSeek-R1-Distill-Qwen-14B":[148],"by":[149],"an":[150],"average":[151],"3.1":[153],"points":[154],"(from":[155],"48.2":[156],"to":[157,165,184],"51.3)":[158],"boosts":[160],"Qwen3-14B":[161],"performance":[162,182],"58.8":[164],"61.1":[166],"six":[168],"benchmarks,":[169],"which":[170],"are":[171],"unseen":[172],"during":[173],"environments.":[178],"Moreover,":[179],"achieves":[181],"comparable":[183],"using":[190],"only":[191],"50":[192],"base":[193],"environments,":[194],"demonstrating":[195],"significant":[196],"efficiency":[197],"in":[198],"utilization.":[200]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-12T00:00:00"}