{"id":"https://openalex.org/W7139913273","doi":"https://doi.org/10.48550/arxiv.2603.19220","title":"Nemotron-Cascade 2: Post-Training LLMs with Cascade RL and Multi-Domain On-Policy Distillation","display_name":"Nemotron-Cascade 2: Post-Training LLMs with Cascade RL and Multi-Domain On-Policy Distillation","publication_year":2026,"publication_date":"2026-03-19","ids":{"openalex":"https://openalex.org/W7139913273","doi":"https://doi.org/10.48550/arxiv.2603.19220"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.19220","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.19220","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.2603.19220","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5130249387","display_name":"Zhuolin Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Zhuolin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130231387","display_name":"Zihan Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Zihan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130229538","display_name":"Yang Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Yang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108499414","display_name":"Wenliang Dai","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dai, Wenliang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130251760","display_name":"Boxin Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Boxin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032699557","display_name":"Sheng-Chieh Lin","orcid":"https://orcid.org/0000-0002-7989-9703"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lin, Sheng-Chieh","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089891838","display_name":"Chankyu Lee","orcid":"https://orcid.org/0000-0002-6933-7459"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lee, Chankyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130236158","display_name":"Yangyi Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Yangyi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130239921","display_name":"Dongfu Jiang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiang, Dongfu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5081773147","display_name":"Jie He","orcid":"https://orcid.org/0000-0002-7136-4181"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"He, Jiafan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130250037","display_name":"Renjie Pi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pi, Renjie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130249895","display_name":"Grace Lam","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lam, Grace","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130251377","display_name":"Nayeon Lee","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lee, Nayeon","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5053672689","display_name":"Alexander Bukharin","orcid":"https://orcid.org/0000-0003-2913-5112"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bukharin, Alexander","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130234664","display_name":"Mohammad Shoeybi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shoeybi, Mohammad","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130251649","display_name":"Bryan Catanzaro","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Catanzaro, Bryan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130228313","display_name":"Wei Ping","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ping, Wei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":17,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.1868000030517578,"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"}},"topics":[{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.1868000030517578,"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":0.16120000183582306,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.13009999692440033,"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/cascade","display_name":"Cascade","score":0.7351999878883362},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.5246999859809875},{"id":"https://openalex.org/keywords/olympiad","display_name":"Olympiad","score":0.45879998803138733},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.4066999852657318},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.3939000070095062},{"id":"https://openalex.org/keywords/distillation","display_name":"Distillation","score":0.36550000309944153}],"concepts":[{"id":"https://openalex.org/C34146451","wikidata":"https://www.wikidata.org/wiki/Q5048094","display_name":"Cascade","level":2,"score":0.7351999878883362},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5390999913215637},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5246999859809875},{"id":"https://openalex.org/C130383907","wikidata":"https://www.wikidata.org/wiki/Q221956","display_name":"Olympiad","level":2,"score":0.45879998803138733},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.4066999852657318},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.39739999175071716},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.3939000070095062},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.36550000309944153},{"id":"https://openalex.org/C191630685","wikidata":"https://www.wikidata.org/wiki/Q4027615","display_name":"Informatics","level":2,"score":0.36010000109672546},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3474999964237213},{"id":"https://openalex.org/C2780428219","wikidata":"https://www.wikidata.org/wiki/Q16952335","display_name":"Cover (algebra)","level":2,"score":0.3305000066757202},{"id":"https://openalex.org/C179518139","wikidata":"https://www.wikidata.org/wiki/Q5140297","display_name":"Coding (social sciences)","level":2,"score":0.3244999945163727},{"id":"https://openalex.org/C2779808786","wikidata":"https://www.wikidata.org/wiki/Q6664603","display_name":"Locality","level":2,"score":0.3149999976158142},{"id":"https://openalex.org/C189430467","wikidata":"https://www.wikidata.org/wiki/Q7293293","display_name":"Ranking (information retrieval)","level":2,"score":0.31220000982284546},{"id":"https://openalex.org/C183115368","wikidata":"https://www.wikidata.org/wiki/Q856577","display_name":"Weighting","level":2,"score":0.30250000953674316}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.19220","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.19220","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.2603.19220","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.19220","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":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"We":[0,147],"introduce":[1,114],"Nemotron-Cascade":[2,80],"2,":[3],"an":[4],"open":[5,35],"30B":[6],"MoE":[7],"model":[8,152],"with":[9,73],"3B":[10],"activated":[11],"parameters":[12],"that":[13,32],"delivers":[14],"best-in-class":[15],"reasoning":[16,29,108],"and":[17,27,63,109,139,154],"strong":[18,141],"agentic":[19,110],"capabilities.":[20],"Despite":[21],"its":[22,25],"compact":[23],"size,":[24],"mathematical":[26],"coding":[28],"performance":[30,49,142],"approaches":[31],"of":[33,107,151],"frontier":[34],"models.":[36],"It":[37],"is":[38],"the":[39,51,57,64,82,119,128,145,149],"second":[40],"open-weight":[41],"LLM,":[42],"after":[43],"DeepSeekV3.2-Speciale-671B-A37B,":[44],"to":[45,79,101,134],"achieve":[46],"Gold":[47],"Medal-level":[48],"in":[50,60],"2025":[52],"International":[53,58],"Mathematical":[54],"Olympiad":[55,59],"(IMO),":[56],"Informatics":[61],"(IOI),":[62],"ICPC":[65],"World":[66],"Finals,":[67],"demonstrating":[68],"remarkably":[69],"high":[70],"intelligence":[71],"density":[72],"20x":[74],"fewer":[75],"parameters.":[76],"In":[77],"contrast":[78],"1,":[81],"key":[83],"technical":[84],"advancements":[85],"are":[86],"as":[87],"follows.":[88],"After":[89],"SFT":[90],"on":[91],"a":[92,103],"meticulously":[93],"curated":[94],"dataset,":[95],"we":[96,113],"substantially":[97],"expand":[98],"Cascade":[99,129],"RL":[100,130],"cover":[102],"much":[104],"broader":[105],"spectrum":[106],"domains.":[111],"Furthermore,":[112],"multi-domain":[115],"on-policy":[116],"distillation":[117],"from":[118],"strongest":[120],"intermediate":[121],"teacher":[122],"models":[123],"for":[124],"each":[125],"domain":[126],"throughout":[127],"process,":[131],"allowing":[132],"us":[133],"efficiently":[135],"recover":[136],"benchmark":[137],"regressions":[138],"sustain":[140],"gains":[143],"along":[144],"way.":[146],"release":[148],"collection":[150],"checkpoint":[153],"training":[155],"data.":[156]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-03-21T00:00:00"}