{"id":"https://openalex.org/W7130534234","doi":"https://doi.org/10.48550/arxiv.2602.16179","title":"EnterpriseBench Corecraft: Training Generalizable Agents on High-Fidelity RL Environments","display_name":"EnterpriseBench Corecraft: Training Generalizable Agents on High-Fidelity RL Environments","publication_year":2026,"publication_date":"2026-02-18","ids":{"openalex":"https://openalex.org/W7130534234","doi":"https://doi.org/10.48550/arxiv.2602.16179"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2602.16179","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.16179","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":"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"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2602.16179","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5123144009","display_name":"Sushant Mehta","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mehta, Sushant","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126372826","display_name":"Logan Ritchie","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ritchie, Logan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054506562","display_name":"Suhaas G. Garre","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Garre, Suhaas","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Niebres, Ian","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Niebres, Ian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Heiner, Nick","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Heiner, Nick","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5126224959","display_name":"Edwin Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Edwin","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":0.17010000348091125,"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":0.17010000348091125,"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/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.15600000321865082,"subfield":{"id":"https://openalex.org/subfields/2718","display_name":"Health Informatics"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},{"id":"https://openalex.org/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.08129999786615372,"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/reinforcement-learning","display_name":"Reinforcement learning","score":0.6815999746322632},{"id":"https://openalex.org/keywords/suite","display_name":"Suite","score":0.6722000241279602},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.6572999954223633},{"id":"https://openalex.org/keywords/workflow","display_name":"Workflow","score":0.5787000060081482},{"id":"https://openalex.org/keywords/rubric","display_name":"Rubric","score":0.5598000288009644},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.4781000018119812},{"id":"https://openalex.org/keywords/ranking","display_name":"Ranking (information retrieval)","score":0.40230000019073486}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7218999862670898},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6815999746322632},{"id":"https://openalex.org/C79581498","wikidata":"https://www.wikidata.org/wiki/Q1367530","display_name":"Suite","level":2,"score":0.6722000241279602},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6572999954223633},{"id":"https://openalex.org/C177212765","wikidata":"https://www.wikidata.org/wiki/Q627335","display_name":"Workflow","level":2,"score":0.5787000060081482},{"id":"https://openalex.org/C111640148","wikidata":"https://www.wikidata.org/wiki/Q847349","display_name":"Rubric","level":2,"score":0.5598000288009644},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5062000155448914},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.4781000018119812},{"id":"https://openalex.org/C189430467","wikidata":"https://www.wikidata.org/wiki/Q7293293","display_name":"Ranking (information retrieval)","level":2,"score":0.40230000019073486},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.39980000257492065},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.36059999465942383},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.3427000045776367},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.33340001106262207},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.3303000032901764},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.3061000108718872},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.2782999873161316},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.2667999863624573},{"id":"https://openalex.org/C184898388","wikidata":"https://www.wikidata.org/wiki/Q1435712","display_name":"Pairwise comparison","level":2,"score":0.2628999948501587},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.25440001487731934}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2602.16179","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.16179","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2602.16179","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.16179","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":"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"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/9","score":0.4033232033252716,"display_name":"Industry, innovation and infrastructure"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"We":[0,19,155],"show":[1],"that":[2,13,70,169,184,192],"training":[3,17],"AI":[4,62],"agents":[5,63],"on":[6,129,142,146,151],"high-fidelity":[7],"reinforcement":[8],"learning":[9],"environments":[10],"produces":[11],"capabilities":[12],"generalize":[14],"beyond":[15],"the":[16,22,66,119,163],"distribution.":[18],"introduce":[20],"CoreCraft,":[21],"first":[23],"environment":[24,158,193],"in":[25],"EnterpriseBench,":[26],"Surge":[27],"AI's":[28],"suite":[29],"of":[30,41,87,117],"agentic":[31],"RL":[32],"environments.":[33],"CoreCraft":[34],"is":[35],"a":[36,42,114],"fully":[37],"operational":[38],"enterprise":[39,182],"simulation":[40],"customer":[43],"support":[44],"organization,":[45],"comprising":[46],"over":[47],"2,500":[48],"entities":[49],"across":[50],"14":[51],"entity":[52],"types":[53],"with":[54,104,162],"23":[55],"unique":[56],"tools,":[57],"designed":[58],"to":[59,124,138],"measure":[60],"whether":[61],"can":[64],"perform":[65],"multi-step,":[67],"domain-specific":[68],"work":[69],"real":[71],"jobs":[72],"demand.":[73],"Frontier":[74],"models":[75],"such":[76],"as":[77],"GPT-5.2":[78],"and":[79,110,149,181,196],"Claude":[80],"Opus":[81],"4.6":[82,103],"solve":[83],"fewer":[84],"than":[85],"30%":[86],"tasks":[88],"when":[89],"all":[90],"expert-authored":[91,175],"rubric":[92],"criteria":[93],"must":[94],"be":[95],"satisfied.":[96],"Using":[97],"this":[98],"environment,":[99],"we":[100],"train":[101],"GLM":[102],"Group":[105],"Relative":[106],"Policy":[107],"Optimization":[108],"(GRPO)":[109],"adaptive":[111],"clipping.":[112],"After":[113],"single":[115],"epoch":[116],"training,":[118],"model":[120],"improves":[121],"from":[122],"25.37%":[123],"36.76%":[125],"task":[126],"pass":[127],"rate":[128],"held-out":[130],"evaluation":[131],"tasks.":[132],"More":[133],"importantly,":[134],"these":[135],"gains":[136],"transfer":[137],"out-of-distribution":[139],"benchmarks:":[140],"+4.5%":[141],"BFCL":[143],"Parallel,":[144],"+7.4%":[145],"Tau2-Bench":[147],"Retail,":[148],"+6.8%":[150],"Tool":[152],"Decathlon":[153],"(Pass@1).":[154],"believe":[156],"three":[157],"properties":[159],"are":[160,198],"consistent":[161],"observed":[164],"transfer:":[165],"task-centric":[166],"world":[167],"building":[168],"optimizes":[170],"for":[171],"diverse,":[172],"challenging":[173],"tasks;":[174],"rubrics":[176],"enabling":[177,201],"reliable":[178],"reward":[179],"computation;":[180],"workflows":[183],"reflect":[185],"realistic":[186],"professional":[187],"patterns.":[188],"Our":[189],"results":[190],"suggest":[191],"quality,":[194],"diversity,":[195],"realism":[197],"key":[199],"factors":[200],"generalizable":[202],"agent":[203],"capabilities.":[204]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-02-20T00:00:00"}