{"id":"https://openalex.org/W4289445724","doi":"https://doi.org/10.48550/arxiv.2207.14502","title":"Language Models Can Teach Themselves to Program Better","display_name":"Language Models Can Teach Themselves to Program Better","publication_year":2022,"publication_date":"2022-07-29","ids":{"openalex":"https://openalex.org/W4289445724","doi":"https://doi.org/10.48550/arxiv.2207.14502"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2207.14502","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2207.14502","pdf_url":"https://arxiv.org/pdf/2207.14502","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/2207.14502","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5014245834","display_name":"Patrick Haluptzok","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Haluptzok, Patrick","raw_affiliation_strings":["Microsoft Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100981228","display_name":"Matthew R. Bowers","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I63966007","display_name":"Massachusetts Institute of Technology","ror":"https://ror.org/042nb2s44","country_code":"US","type":"education","lineage":["https://openalex.org/I63966007"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bowers, Matthew","raw_affiliation_strings":["MIT"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"MIT","institution_ids":["https://openalex.org/I63966007"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5086904495","display_name":"Adam Tauman Kalai","orcid":"https://orcid.org/0000-0002-4559-8574","observed_orcids":["https://orcid.org/0000-0002-4559-8574"]},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kalai, Adam Tauman","raw_affiliation_strings":["Microsoft Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research","institution_ids":["https://openalex.org/I1290206253"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":18,"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/T10260","display_name":"Software Engineering Research","score":0.7080000042915344,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10260","display_name":"Software Engineering Research","score":0.7080000042915344,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10533","display_name":"Teaching and Learning Programming","score":0.1923000067472458,"subfield":{"id":"https://openalex.org/subfields/1706","display_name":"Computer Science Applications"},"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":0.04809999838471413,"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/self-play","display_name":"self-play","score":0.6439999938011169},{"id":"https://openalex.org/keywords/code-generation","display_name":"code generation","score":0.7269999980926514},{"id":"https://openalex.org/keywords/programming-puzzles","display_name":"programming puzzles","score":0.5619999766349792},{"id":"https://openalex.org/keywords/fine-tuning","display_name":"fine-tuning","score":0.6549999713897705},{"id":"https://openalex.org/keywords/code-llms","display_name":"code LLMs","score":0.5350000262260437}],"concepts":[{"id":"https://openalex.org/C55439883","wikidata":"https://www.wikidata.org/wiki/Q360812","display_name":"Correctness","level":2,"score":0.854010283946991},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8517047166824341},{"id":"https://openalex.org/C122783720","wikidata":"https://www.wikidata.org/wiki/Q183065","display_name":"Interpreter","level":2,"score":0.8481079936027527},{"id":"https://openalex.org/C519991488","wikidata":"https://www.wikidata.org/wiki/Q28865","display_name":"Python (programming language)","level":2,"score":0.8376374244689941},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.7406221628189087},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.45158228278160095},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2207.14502","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2207.14502","pdf_url":"https://arxiv.org/pdf/2207.14502","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.2207.14502","doi":"https://doi.org/10.48550/arxiv.2207.14502","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2207.14502","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2207.14502","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2207.14502","pdf_url":"https://arxiv.org/pdf/2207.14502","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/W4415185565","https://openalex.org/W4365205411","https://openalex.org/W4417163889","https://openalex.org/W4403884304","https://openalex.org/W4361807047","https://openalex.org/W7117141836","https://openalex.org/W7151369811","https://openalex.org/W4396214504","https://openalex.org/W7161723905","https://openalex.org/W4368755703"],"abstract_inverted_index":{"Recent":[0],"Language":[1],"Models":[2],"(LMs)":[3],"achieve":[4],"breakthrough":[5],"performance":[6,75],"in":[7,24],"code":[8,143],"generation":[9],"when":[10,81],"trained":[11],"on":[12,85,129],"human-authored":[13],"problems,":[14],"even":[15],"solving":[16],"some":[17],"competitive-programming":[18],"problems.":[19],"Self-play":[20],"has":[21],"proven":[22],"useful":[23],"games":[25],"such":[26],"as":[27,106],"Go,":[28],"and":[29,62,90,152],"thus":[30,93],"it":[31,52,82],"is":[32,53,76,83],"natural":[33],"to":[34,45,58,79,148],"ask":[35],"whether":[36],"LMs":[37],"can":[38,119],"generate":[39,149],"their":[40,47,154],"own":[41,87,155],"instructive":[42,150],"programming":[43,60,107],"problems":[44,61,89,151],"improve":[46,80,153],"performance.":[48,156],"We":[49],"show":[50],"that":[51],"possible":[54],"for":[55,67,123,142],"an":[56,146],"LM":[57],"synthesize":[59],"solutions,":[63],"which":[64],"are":[65,103],"filtered":[66],"correctness":[68,124],"by":[69,125],"a":[70,113],"Python":[71,100],"interpreter.":[72,101],"The":[73],"LM's":[74],"then":[77],"seen":[78],"fine-tuned":[84],"its":[86],"synthetic":[88],"verified":[91,122],"solutions;":[92],"the":[94,99,140],"model":[95],"'improves":[96],"itself'":[97],"using":[98],"Problems":[102],"specified":[104],"formally":[105],"puzzles":[108],"[Schuster":[109],"et":[110],"al.,":[111],"2021],":[112],"code-based":[114],"problem":[115],"format":[116],"where":[117],"solutions":[118],"easily":[120],"be":[121],"execution.":[126],"In":[127],"experiments":[128],"publicly-available":[130],"LMs,":[131,144],"test":[132],"accuracy":[133],"more":[134],"than":[135],"doubles.":[136],"This":[137],"work":[138],"demonstrates":[139],"potential":[141],"with":[145],"interpreter,":[147]},"counts_by_year":[{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":8},{"year":2023,"cited_by_count":7}],"updated_date":"2026-10-16T09:59:22.235173","created_date":"2022-08-02T00:00:00"}