{"id":"https://openalex.org/W4414850267","doi":"https://doi.org/10.48550/arxiv.2502.17387","title":"Big-Math: A Large-Scale, High-Quality Math Dataset for Reinforcement Learning in Language Models","display_name":"Big-Math: A Large-Scale, High-Quality Math Dataset for Reinforcement Learning in Language Models","publication_year":2025,"publication_date":"2025-02-24","ids":{"openalex":"https://openalex.org/W4414850267","doi":"https://doi.org/10.48550/arxiv.2502.17387"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2502.17387","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2502.17387","pdf_url":"https://arxiv.org/pdf/2502.17387","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2502.17387","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5090026363","display_name":"Alon Albalak","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Albalak, Alon","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003505678","display_name":"Duy Phung","orcid":"https://orcid.org/0009-0006-0199-6016","observed_orcids":["https://orcid.org/0009-0006-0199-6016"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Phung, Duy","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5093931915","display_name":"Nathan Lile","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lile, Nathan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028522433","display_name":"Rafael Rafailov","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rafailov, Rafael","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085625870","display_name":"Kanishk Gandhi","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gandhi, Kanishk","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5004714414","display_name":"Louis Castricato","orcid":"https://orcid.org/0000-0003-2996-886X","observed_orcids":["https://orcid.org/0000-0003-2996-886X"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Castricato, Louis","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5088818925","display_name":"Anikait Singh","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Singh, Anikait","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Blagden, Chase","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Blagden, Chase","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023185054","display_name":"Violet Xiang","orcid":"https://orcid.org/0000-0002-1584-6231","observed_orcids":["https://orcid.org/0000-0002-1584-6231"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiang, Violet","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5094363682","display_name":"Dakota Mahan","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mahan, Dakota","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5069105490","display_name":"Nick Haber","orcid":"https://orcid.org/0000-0001-8804-7804","observed_orcids":["https://orcid.org/0000-0001-8804-7804"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Haber, Nick","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":0.43479999899864197,"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":0.43479999899864197,"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/T13523","display_name":"Mathematics, Computing, and Information Processing","score":0.10719999670982361,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/T12522","display_name":"Mathematics Education and Pedagogy","score":0.08760000020265579,"subfield":{"id":"https://openalex.org/subfields/2604","display_name":"Applied Mathematics"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/mathematical-reasoning","display_name":"mathematical reasoning","score":0.6370000243186951},{"id":"https://openalex.org/keywords/data-curation","display_name":"data curation","score":0.6800000071525574},{"id":"https://openalex.org/keywords/multiple-choice-questions","display_name":"multiple-choice questions","score":0.671999990940094},{"id":"https://openalex.org/keywords/analytical-solution","display_name":"analytical solution","score":0.8309999704360962},{"id":"https://openalex.org/keywords/gsm8k","display_name":"GSM8K","score":0.7490000128746033},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.6169999837875366}],"concepts":[{"id":"https://openalex.org/C85847156","wikidata":"https://www.wikidata.org/wiki/Q59015987","display_name":"Verifiable secret sharing","level":3,"score":0.7544999718666077},{"id":"https://openalex.org/C174348530","wikidata":"https://www.wikidata.org/wiki/Q188635","display_name":"Bridging (networking)","level":2,"score":0.7031999826431274},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6398000121116638},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.5843999981880188},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.5403000116348267},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.47530001401901245},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46639999747276306},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.40860000252723694},{"id":"https://openalex.org/C197115733","wikidata":"https://www.wikidata.org/wiki/Q1003136","display_name":"Forcing (mathematics)","level":2,"score":0.3659999966621399},{"id":"https://openalex.org/C33676613","wikidata":"https://www.wikidata.org/wiki/Q13415176","display_name":"Dimension (graph theory)","level":2,"score":0.31349998712539673},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.31290000677108765},{"id":"https://openalex.org/C2780966255","wikidata":"https://www.wikidata.org/wiki/Q5474306","display_name":"Foundation (evidence)","level":2,"score":0.3084999918937683},{"id":"https://openalex.org/C133462117","wikidata":"https://www.wikidata.org/wiki/Q4929239","display_name":"Data collection","level":2,"score":0.296999990940094},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.28349998593330383},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.2619999945163727},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.2565999925136566}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2502.17387","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2502.17387","pdf_url":"https://arxiv.org/pdf/2502.17387","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2502.17387","doi":"https://doi.org/10.48550/arxiv.2502.17387","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2502.17387","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":"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2502.17387","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2502.17387","pdf_url":"https://arxiv.org/pdf/2502.17387","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"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":["https://openalex.org/W4416052877","https://openalex.org/W4403928588","https://openalex.org/W4392538380","https://openalex.org/W7141746640","https://openalex.org/W4416246779","https://openalex.org/W4409348209","https://openalex.org/W4412945228","https://openalex.org/W7155420302","https://openalex.org/W4402670720","https://openalex.org/W4410348938"],"abstract_inverted_index":{"Increasing":[0],"interest":[1],"in":[2,122,258],"reasoning":[3,257],"models":[4,234],"has":[5],"led":[6],"math":[7,22,63,171],"to":[8,45,162],"become":[9],"a":[10,26,34,57,108,157,200,210,220,227,252],"prominent":[11],"testing":[12],"ground":[13],"for":[14,70,170,195,233,255],"algorithmic":[15],"and":[16,49,81,105,174,218,238,248],"methodological":[17],"improvements.":[18],"However,":[19],"existing":[20,167],"open":[21],"datasets":[23,169],"either":[24],"contain":[25],"small":[27],"collection":[28],"of":[29,37,40,59,115,180,203,213,223,230,235],"high-quality,":[30],"human-written":[31],"problems":[32,39,94,100,106],"or":[33],"large":[35],"corpus":[36],"machine-generated":[38],"uncertain":[41],"quality,":[42],"forcing":[43],"researchers":[44],"choose":[46],"between":[47,245],"quality":[48,114,247],"quantity.":[50],"In":[51],"this":[52],"work,":[53],"we":[54,77,117,134,189],"present":[55],"Big-Math,":[56,76,116],"dataset":[58],"over":[60],"250,000":[61],"high-quality":[62],"questions":[64,87,138,144,155,192],"with":[65,95,107,139],"verifiable":[66,97],"answers,":[67,141],"purposefully":[68],"made":[69],"reinforcement":[71],"learning":[72],"(RL).":[73],"To":[74,111],"create":[75],"rigorously":[78],"filter,":[79],"clean,":[80],"curate":[82],"openly":[83],"available":[84],"datasets,":[85],"extracting":[86],"that":[88,101,149,188,207],"satisfy":[89],"our":[90,123,131,184],"three":[91],"desiderata:":[92],"(1)":[93],"uniquely":[96],"solutions,":[98],"(2)":[99],"are":[102],"open-ended,":[103],"(3)":[104],"closed-form":[109],"solution.":[110],"ensure":[112],"the":[113,128,163,191,204,243],"manually":[118],"verify":[119],"each":[120],"step":[121],"filtering":[124,132,186],"process.":[125],"Based":[126],"on":[127],"findings":[129],"from":[130],"process,":[133],"introduce":[135],"47,000":[136],"new":[137],"verified":[140],"Big-Math-Reformulated:":[142],"closed-ended":[143],"(i.e.":[145],"multiple":[146],"choice":[147],"questions)":[148],"have":[150],"been":[151],"reformulated":[152],"as":[153],"open-ended":[154],"through":[156],"systematic":[158],"reformulation":[159],"algorithm.":[160],"Compared":[161],"most":[164,193],"commonly":[165],"used":[166],"open-source":[168],"reasoning,":[172],"GSM8k":[173],"MATH,":[175],"Big-Math":[176,208,250],"is":[177],"an":[178],"order":[179],"magnitude":[181],"larger,":[182],"while":[183],"rigorous":[185,201],"ensures":[187],"maintain":[190],"suitable":[194],"RL.":[196],"We":[197],"also":[198],"provide":[199],"analysis":[202],"dataset,":[205],"finding":[206],"contains":[209],"high":[211],"degree":[212],"diversity":[214],"across":[215],"problem":[216,224],"domains,":[217],"incorporates":[219],"wide":[221,228],"range":[222,229],"difficulties,":[225],"enabling":[226],"downstream":[231],"uses":[232],"varying":[236],"capabilities":[237],"training":[239],"requirements.":[240],"By":[241],"bridging":[242],"gap":[244],"data":[246],"quantity,":[249],"establish":[251],"robust":[253],"foundation":[254],"advancing":[256],"LLMs.":[259]},"counts_by_year":[],"updated_date":"2026-10-13T10:08:12.033832","created_date":"2025-10-10T00:00:00"}