{"id":"https://openalex.org/W4407806895","doi":"https://doi.org/10.48550/arxiv.2502.14499","title":"MLGym: A New Framework and Benchmark for Advancing AI Research Agents","display_name":"MLGym: A New Framework and Benchmark for Advancing AI Research Agents","publication_year":2025,"publication_date":"2025-02-20","ids":{"openalex":"https://openalex.org/W4407806895","doi":"https://doi.org/10.48550/arxiv.2502.14499"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2502.14499","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2502.14499","pdf_url":"https://arxiv.org/pdf/2502.14499","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,"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"},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2502.14499","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5050963493","display_name":"Deepak Nathani","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nathani, Deepak","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5099173851","display_name":"Lovish Madaan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Madaan, Lovish","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112494018","display_name":"Nicholas Roberts","orcid":"https://orcid.org/0009-0003-4762-2202"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Roberts, Nicholas","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5116361685","display_name":"Nikolay Bashlykov","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bashlykov, Nikolay","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109640343","display_name":"Ajay Menon","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Menon, Ajay","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015148245","display_name":"Vincent Moens","orcid":"https://orcid.org/0000-0002-5810-086X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Moens, Vincent","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042977938","display_name":"Amar Budhiraja","orcid":"https://orcid.org/0000-0001-5891-3759"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Budhiraja, Amar","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020439365","display_name":"Despoina Magka","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Magka, Despoina","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5116361686","display_name":"Vladislav Vorotilov","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Vorotilov, Vladislav","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091730595","display_name":"Gaurav Chaurasia","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chaurasia, Gaurav","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018069499","display_name":"Dieuwke Hupkes","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hupkes, Dieuwke","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5031282729","display_name":"Ricardo Silveira Cabral","orcid":"https://orcid.org/0000-0002-5471-1778"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cabral, Ricardo Silveira","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5079740967","display_name":"Tatiana Shavrina","orcid":"https://orcid.org/0000-0002-6976-0185"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shavrina, Tatiana","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059094093","display_name":"Jakob Foerster","orcid":"https://orcid.org/0000-0001-9688-2498"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Foerster, Jakob","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062949033","display_name":"Yoram Bachrach","orcid":"https://orcid.org/0000-0002-4382-7636"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bachrach, Yoram","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100702485","display_name":"William Yang Wang","orcid":"https://orcid.org/0000-0001-6153-8240"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, William Yang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5018702533","display_name":"Roberta R\u0103ileanu","orcid":"https://orcid.org/0009-0003-2057-2903"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Raileanu, Roberta","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":4,"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/T10215","display_name":"Semantic Web and Ontologies","score":0.661300003528595,"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/T10215","display_name":"Semantic Web and Ontologies","score":0.661300003528595,"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/T10906","display_name":"AI-based Problem Solving and Planning","score":0.5651999711990356,"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/T11986","display_name":"Scientific Computing and Data Management","score":0.5618000030517578,"subfield":{"id":"https://openalex.org/subfields/1802","display_name":"Information Systems and Management"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.8130365610122681},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.45254963636398315},{"id":"https://openalex.org/keywords/data-science","display_name":"Data science","score":0.33310842514038086},{"id":"https://openalex.org/keywords/geography","display_name":"Geography","score":0.12503182888031006},{"id":"https://openalex.org/keywords/cartography","display_name":"Cartography","score":0.07846406102180481}],"concepts":[{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.8130365610122681},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.45254963636398315},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.33310842514038086},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.12503182888031006},{"id":"https://openalex.org/C58640448","wikidata":"https://www.wikidata.org/wiki/Q42515","display_name":"Cartography","level":1,"score":0.07846406102180481}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2502.14499","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2502.14499","pdf_url":"https://arxiv.org/pdf/2502.14499","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,"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.2502.14499","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2502.14499","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,"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":"pmh:oai:arXiv.org:2502.14499","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2502.14499","pdf_url":"https://arxiv.org/pdf/2502.14499","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,"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":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2378211422","https://openalex.org/W4321353415","https://openalex.org/W2745001401","https://openalex.org/W2130974462","https://openalex.org/W2028665553","https://openalex.org/W2086519370","https://openalex.org/W4246352526"],"abstract_inverted_index":{"We":[0,108,166,194],"introduce":[1],"Meta":[2],"MLGym":[3,132],"and":[4,9,13,48,65,81,84,97,128,142,198],"MLGym-Bench,":[5],"a":[6,105,110],"new":[7,79,139,156],"framework":[8,133,197],"benchmark":[10,199],"for":[11,27,39,159],"evaluating":[12],"developing":[14],"LLM":[15,211],"agents":[16,161],"on":[17,34,104,118,162,174],"AI":[18,50,73,163,207],"research":[19,33,51,74,164,203,208],"tasks.":[20,165],"This":[21],"is":[22],"the":[23,95,175,206],"first":[24],"Gym":[25],"environment":[26],"machine":[28],"learning":[29,36,157],"(ML)":[30],"tasks,":[31,140],"enabling":[32],"reinforcement":[35,63],"(RL)":[37],"algorithms":[38,158],"training":[40,90,160],"such":[41,56,76,121],"agents.":[42,212],"MLGym-bench":[43],"consists":[44],"of":[45,112,210],"13":[46],"diverse":[47,54],"open-ended":[49],"tasks":[52,70],"from":[53],"domains":[55],"as":[57,77,122,152,154],"computer":[58],"vision,":[59],"natural":[60],"language":[61,115],"processing,":[62],"learning,":[64],"game":[66],"theory.":[67],"Solving":[68],"these":[69],"requires":[71],"real-world":[72],"skills":[75],"generating":[78],"ideas":[80],"hypotheses,":[82,188],"creating":[83],"processing":[85],"data,":[86],"implementing":[87],"ML":[88],"methods,":[89],"models,":[91],"running":[92],"experiments,":[93],"analyzing":[94],"results,":[96],"iterating":[98],"through":[99],"this":[100],"process":[101],"to":[102,137,200],"improve":[103,173],"given":[106,176],"task.":[107],"evaluate":[109,143],"number":[111],"frontier":[113,170],"large":[114],"models":[116,144,171],"(LLMs)":[117],"our":[119,196],"benchmarks":[120],"Claude-3.5-Sonnet,":[123],"Llama-3.1":[124],"405B,":[125],"GPT-4o,":[126],"o1-preview,":[127],"Gemini-1.5":[129],"Pro.":[130],"Our":[131],"makes":[134],"it":[135],"easy":[136],"add":[138],"integrate":[141],"or":[145,191],"agents,":[146],"generate":[147,186],"synthetic":[148],"data":[149],"at":[150],"scale,":[151],"well":[153],"develop":[155],"find":[167],"that":[168],"current":[169],"can":[172],"baselines,":[177],"usually":[178],"by":[179],"finding":[180],"better":[181],"hyperparameters,":[182],"but":[183],"do":[184],"not":[185],"novel":[187],"algorithms,":[189],"architectures,":[190],"substantial":[192],"improvements.":[193],"open-source":[195],"facilitate":[201],"future":[202],"in":[204],"advancing":[205],"capabilities":[209]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":2}],"updated_date":"2026-08-26T07:47:46.906454","created_date":"2025-10-10T00:00:00"}