{"id":"https://openalex.org/W7162805565","doi":"https://doi.org/10.48550/arxiv.2605.28829","title":"Aryabhata 2: Scaling Reinforcement Learning for Advanced STEM Reasoning","display_name":"Aryabhata 2: Scaling Reinforcement Learning for Advanced STEM Reasoning","publication_year":2026,"publication_date":"2026-04-10","ids":{"openalex":"https://openalex.org/W7162805565","doi":"https://doi.org/10.48550/arxiv.2605.28829"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.28829","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.28829","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":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.28829","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5137382339","display_name":"Ritvik Rastogi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rastogi, Ritvik","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137346705","display_name":"Vishal Singh","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Singh, Vishal","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137369721","display_name":"Tejas Chaudhari","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chaudhari, Tejas","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5032451828","display_name":"Sandeep Varma","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Varma, Sandeep","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":4,"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/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.24079999327659607,"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/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.24079999327659607,"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":0.09969999641180038,"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/T10028","display_name":"Topic Modeling","score":0.07729999721050262,"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":0.7257000207901001},{"id":"https://openalex.org/keywords/construct","display_name":"Construct (python library)","score":0.5648000240325928},{"id":"https://openalex.org/keywords/verifiable-secret-sharing","display_name":"Verifiable secret sharing","score":0.3959999978542328},{"id":"https://openalex.org/keywords/knowledge-base","display_name":"Knowledge base","score":0.37869998812675476},{"id":"https://openalex.org/keywords/base","display_name":"Base (topology)","score":0.31450000405311584}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7257000207901001},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6865000128746033},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6085000038146973},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.5648000240325928},{"id":"https://openalex.org/C85847156","wikidata":"https://www.wikidata.org/wiki/Q59015987","display_name":"Verifiable secret sharing","level":3,"score":0.3959999978542328},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3828999996185303},{"id":"https://openalex.org/C4554734","wikidata":"https://www.wikidata.org/wiki/Q593744","display_name":"Knowledge base","level":2,"score":0.37869998812675476},{"id":"https://openalex.org/C42058472","wikidata":"https://www.wikidata.org/wiki/Q810214","display_name":"Base (topology)","level":2,"score":0.31450000405311584},{"id":"https://openalex.org/C47177190","wikidata":"https://www.wikidata.org/wiki/Q207137","display_name":"Curriculum","level":2,"score":0.2874000072479248},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.2718999981880188},{"id":"https://openalex.org/C58546491","wikidata":"https://www.wikidata.org/wiki/Q1150207","display_name":"Competitive advantage","level":2,"score":0.25870001316070557}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.28829","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.28829","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":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.28829","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.28829","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":"article"},"sustainable_development_goals":[{"score":0.654571533203125,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Competitive":[0],"STEM":[1,63,145],"examinations":[2],"such":[3,124],"as":[4,118,120,125],"JEE":[5,112,114],"and":[6,15,22,80,116,131],"NEET":[7],"require":[8],"multi-step":[9],"symbolic":[10],"reasoning,":[11],"precise":[12],"numerical":[13],"computation,":[14],"deep":[16],"conceptual":[17],"understanding":[18],"across":[19],"physics,":[20],"chemistry,":[21],"mathematics.":[23],"Recent":[24],"large":[25],"language":[26,59],"models":[27],"perform":[28],"strongly":[29],"on":[30,107,143],"common":[31],"reasoning":[32,122,146],"benchmarks,":[33,110],"yet":[34],"they":[35],"remain":[36],"difficult":[37],"to":[38,154],"deploy":[39],"at":[40],"scale,":[41],"where":[42],"millions":[43],"of":[44],"student":[45],"doubts":[46],"demand":[47],"domain-specific,":[48],"consistently":[49],"structured":[50],"problem":[51],"solving.":[52],"We":[53,103],"introduce":[54],"Aryabhata":[55,105,136],"2,":[56],"a":[57,76],"reasoning-focused":[58],"model":[60,141],"for":[61],"competitive":[62,108,144],"examinations,":[64],"trained":[65],"via":[66,97],"reinforcement-learning":[67],"post-training.":[68],"Using":[69],"PhysicsWallah's":[70],"internal":[71],"question":[72],"banks,":[73],"we":[74],"construct":[75],"high-quality":[77],"training":[78],"curriculum":[79],"post-train":[81],"GPT-OSS-20B":[82,142],"through":[83],"reinforcement":[84,92],"learning":[85,93],"with":[86,94],"verifiable":[87],"rewards.":[88],"Training":[89],"combines":[90],"prolonged":[91],"broadened":[95],"exploration":[96],"progressively":[98],"larger":[99],"rollout":[100],"group":[101],"sizes.":[102],"evaluate":[104],"2":[106,137],"examination":[109],"including":[111],"Main,":[113],"Advanced,":[115],"NEET,":[117],"well":[119],"out-of-distribution":[121],"datasets":[123],"AIME,":[126],"HMMT,":[127],"MMLU-Pro,":[128],"MMLU-Redux":[129],"2.0,":[130],"GPQA.":[132],"Results":[133],"show":[134],"that":[135],"outperforms":[138],"its":[139],"base":[140],"while":[147],"requiring":[148],"substantially":[149],"fewer":[150],"output":[151],"tokens":[152],"(up":[153],"64\\%":[155],"fewer).":[156]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-05-30T00:00:00"}