{"id":"https://openalex.org/W4388185044","doi":"https://doi.org/10.48550/arxiv.2310.20689","title":"Learning From Mistakes Makes LLM Better Reasoner","display_name":"Learning From Mistakes Makes LLM Better Reasoner","publication_year":2023,"publication_date":"2023-10-31","ids":{"openalex":"https://openalex.org/W4388185044","doi":"https://doi.org/10.48550/arxiv.2310.20689"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2310.20689","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2310.20689","pdf_url":"https://arxiv.org/pdf/2310.20689","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/2310.20689","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5027673336","display_name":"Shengnan An","orcid":"https://orcid.org/0009-0003-4989-4607","observed_orcids":["https://orcid.org/0009-0003-4989-4607"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"An, Shengnan","raw_affiliation_strings":["Jiaotong University ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Jiaotong University ,","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102574240","display_name":"Zexiong Ma","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ma, Zexiong","raw_affiliation_strings":["Jiaotong University ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Jiaotong University ,","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5061546771","display_name":"Zeqi Lin","orcid":"https://orcid.org/0009-0001-2563-7601","observed_orcids":["https://orcid.org/0009-0001-2563-7601"]},"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":"Lin, Zeqi","raw_affiliation_strings":["\u2663 Microsoft Corporation ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"\u2663 Microsoft Corporation ,","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5047405956","display_name":"Nanning Zheng","orcid":"https://orcid.org/0000-0003-1608-8257","observed_orcids":["https://orcid.org/0000-0003-1608-8257"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zheng, Nanning","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025118710","display_name":"Jian\u2013Guang Lou","orcid":"https://orcid.org/0000-0001-8496-033X","observed_orcids":["https://orcid.org/0000-0001-8496-033X"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lou, Jian-Guang","raw_affiliation_strings":["Jiaotong University ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Jiaotong University ,","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5051745436","display_name":"Weizhu Chen","orcid":"https://orcid.org/0000-0003-4196-7229","observed_orcids":["https://orcid.org/0000-0003-4196-7229"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Weizhu","raw_affiliation_strings":["Jiaotong University ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Jiaotong University ,","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":7,"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":0.7778000235557556,"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":0.7778000235557556,"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":0.0494999997317791,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.04879999905824661,"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/large-language-models","display_name":"large language models","score":0.8489999771118164},{"id":"https://openalex.org/keywords/llm-reasoning","display_name":"LLM reasoning","score":0.6539999842643738},{"id":"https://openalex.org/keywords/learning-from-errors","display_name":"learning from mistakes","score":0.9010000228881836},{"id":"https://openalex.org/keywords/chain-of-thought-fine-tuning","display_name":"chain-of-thought fine-tuning","score":0.8090000152587891},{"id":"https://openalex.org/keywords/mathematical-problem-solving","display_name":"mathematical problem solving","score":0.7760000228881836},{"id":"https://openalex.org/keywords/error-driven-learning","display_name":"error-based learning","score":0.5529999732971191}],"concepts":[{"id":"https://openalex.org/C9616225","wikidata":"https://www.wikidata.org/wiki/Q3929429","display_name":"Semantic reasoner","level":2,"score":0.9286485910415649},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.47138458490371704},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.36610227823257446},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3359348177909851}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2310.20689","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2310.20689","pdf_url":"https://arxiv.org/pdf/2310.20689","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.2310.20689","doi":"https://doi.org/10.48550/arxiv.2310.20689","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2310.20689","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:2310.20689","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2310.20689","pdf_url":"https://arxiv.org/pdf/2310.20689","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":[{"id":"https://openalex.org/G3833898155","display_name":null,"funder_award_id":"62088102","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4388185044.pdf","grobid_xml":"https://content.openalex.org/works/W4388185044.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4402672106","https://openalex.org/W4404781579","https://openalex.org/W4412888241","https://openalex.org/W4416035345","https://openalex.org/W7219090560","https://openalex.org/W7116869808","https://openalex.org/W4393335458","https://openalex.org/W7168466332","https://openalex.org/W4412944855","https://openalex.org/W7213301124"],"abstract_inverted_index":{"Large":[0],"language":[1],"models":[2,173],"(LLMs)":[3],"recently":[4],"exhibited":[5],"remarkable":[6],"reasoning":[7,17,78,132],"capabilities":[8],"on":[9,146],"solving":[10],"math":[11,44],"problems.":[12],"To":[13],"further":[14,142],"improve":[15,165],"their":[16,169],"capabilities,":[18],"this":[19,61],"work":[20],"explores":[21],"whether":[22],"LLMs":[23,130,163],"can":[24],"LEarn":[25],"from":[26,49,80,168],"MistAkes":[27],"(LEMA),":[28],"akin":[29],"to":[30,41,57,90,164],"the":[31,92,96,99,102,106,120,147],"human":[32,37],"learning":[33,63,167],"process.":[34],"Consider":[35],"a":[36,43,88,113,159],"student":[38],"who":[39],"failed":[40],"solve":[42],"problem,":[45],"he":[46,52],"will":[47],"learn":[48],"what":[50],"mistake":[51,93,103],"has":[53],"made":[54],"and":[55,83,104,131,153,174],"how":[56],"correct":[58,101],"it.":[59],"Mimicking":[60],"error-driven":[62],"process,":[64],"LEMA":[65,136],"incorporates":[66],"mistake-correction":[67],"data":[68,152],"pairs":[69],"during":[70],"fine-tuning":[71],"LLMs.":[72],"Specifically,":[73],"we":[74,111],"first":[75],"collect":[76],"inaccurate":[77],"paths":[79],"various":[81,129],"LLMs,":[82],"then":[84],"employ":[85],"GPT-4":[86],"as":[87],"''corrector''":[89],"identify":[91],"step,":[94],"explain":[95],"reason":[97],"for":[98,123,162],"mistake,":[100],"generate":[105],"final":[107],"answer.":[108],"In":[109],"addition,":[110],"apply":[112],"correction-centric":[114],"evolution":[115],"strategy":[116],"that":[117,135],"effectively":[118,137],"expands":[119],"question":[121],"set":[122],"generating":[124],"correction":[125,154],"data.":[126,155],"Experiments":[127],"across":[128],"tasks":[133],"show":[134],"improves":[138],"CoT-alone":[139],"fine-tuning.":[140],"Our":[141,171],"ablations":[143],"shed":[144],"light":[145],"non-homogeneous":[148],"effectiveness":[149],"between":[150],"CoT":[151],"These":[156],"results":[157],"suggest":[158],"significant":[160],"potential":[161],"through":[166],"mistakes.":[170],"code,":[172],"prompts":[175],"are":[176],"publicly":[177],"available":[178],"at":[179],"https://github.com/microsoft/LEMA.":[180]},"counts_by_year":[{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":4}],"updated_date":"2026-10-16T09:59:22.235173","created_date":"2023-11-02T00:00:00"}