{"id":"https://openalex.org/W6929447323","doi":"https://doi.org/10.48550/arxiv.2504.14523","title":"Learning from Reasoning Failures via Synthetic Data Generation","display_name":"Learning from Reasoning Failures via Synthetic Data Generation","publication_year":2025,"publication_date":"2025-04-20","ids":{"openalex":"https://openalex.org/W6929447323","doi":"https://doi.org/10.48550/arxiv.2504.14523"},"language":"en","primary_location":{"id":"doi:10.48550/arxiv.2504.14523","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2504.14523","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":"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"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2504.14523","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Stan, Gabriela Ben Melech","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Stan, Gabriela Ben Melech","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Aflalo, Estelle","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Aflalo, Estelle","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Madasu, Avinash","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Madasu, Avinash","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Lal, Vasudev","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lal, Vasudev","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":null,"display_name":"Howard, Phillip","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Howard, Phillip","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":1,"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":true,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.6108999848365784,"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":0.6108999848365784,"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":0.1737000048160553,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.08259999752044678,"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/synthetic-data","display_name":"Synthetic data","score":0.6577000021934509},{"id":"https://openalex.org/keywords/variety","display_name":"Variety (cybernetics)","score":0.39160001277923584},{"id":"https://openalex.org/keywords/generative-grammar","display_name":"Generative grammar","score":0.383899986743927},{"id":"https://openalex.org/keywords/data-type","display_name":"Data type","score":0.3481000065803528},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.32659998536109924},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.3228999972343445},{"id":"https://openalex.org/keywords/data-modeling","display_name":"Data modeling","score":0.29159998893737793},{"id":"https://openalex.org/keywords/generative-model","display_name":"Generative model","score":0.2915000021457672}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7287999987602234},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.6765000224113464},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.6577000021934509},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6208999752998352},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.39160001277923584},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.383899986743927},{"id":"https://openalex.org/C138958017","wikidata":"https://www.wikidata.org/wiki/Q190087","display_name":"Data type","level":2,"score":0.3481000065803528},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.32659998536109924},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3228999972343445},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2957000136375427},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.29159998893737793},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.2915000021457672},{"id":"https://openalex.org/C86827895","wikidata":"https://www.wikidata.org/wiki/Q7098582","display_name":"Opportunistic reasoning","level":4,"score":0.24899999797344208},{"id":"https://openalex.org/C20162079","wikidata":"https://www.wikidata.org/wiki/Q1151406","display_name":"Case-based reasoning","level":2,"score":0.2467000037431717},{"id":"https://openalex.org/C109747225","wikidata":"https://www.wikidata.org/wiki/Q815758","display_name":"Scarcity","level":2,"score":0.2425999939441681},{"id":"https://openalex.org/C37335422","wikidata":"https://www.wikidata.org/wiki/Q6888134","display_name":"Model-based reasoning","level":3,"score":0.23160000145435333},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.22930000722408295},{"id":"https://openalex.org/C159032336","wikidata":"https://www.wikidata.org/wiki/Q2488768","display_name":"Non-monotonic logic","level":2,"score":0.2289000004529953},{"id":"https://openalex.org/C146849305","wikidata":"https://www.wikidata.org/wiki/Q370766","display_name":"Ground truth","level":2,"score":0.22589999437332153},{"id":"https://openalex.org/C89288958","wikidata":"https://www.wikidata.org/wiki/Q7301504","display_name":"Reasoning system","level":2,"score":0.2092999964952469},{"id":"https://openalex.org/C2776145971","wikidata":"https://www.wikidata.org/wiki/Q30673951","display_name":"Labeled data","level":2,"score":0.20340000092983246},{"id":"https://openalex.org/C2776291640","wikidata":"https://www.wikidata.org/wiki/Q2912517","display_name":"Value (mathematics)","level":2,"score":0.20110000669956207},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.2003999948501587},{"id":"https://openalex.org/C43971567","wikidata":"https://www.wikidata.org/wiki/Q3142865","display_name":"Logical reasoning","level":2,"score":0.2003999948501587},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.19779999554157257},{"id":"https://openalex.org/C110332635","wikidata":"https://www.wikidata.org/wiki/Q629498","display_name":"Genetic programming","level":2,"score":0.1973000019788742},{"id":"https://openalex.org/C2780977526","wikidata":"https://www.wikidata.org/wiki/Q42417149","display_name":"Data exploration","level":3,"score":0.1923000067472458},{"id":"https://openalex.org/C3020493868","wikidata":"https://www.wikidata.org/wiki/Q55631277","display_name":"Real world data","level":2,"score":0.19120000302791595},{"id":"https://openalex.org/C2779714256","wikidata":"https://www.wikidata.org/wiki/Q25305062","display_name":"Multiple Models","level":2,"score":0.19089999794960022},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.18870000541210175},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.18410000205039978},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.18400000035762787},{"id":"https://openalex.org/C55037315","wikidata":"https://www.wikidata.org/wiki/Q5421151","display_name":"Experimental data","level":2,"score":0.18310000002384186},{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.18289999663829803},{"id":"https://openalex.org/C2776937632","wikidata":"https://www.wikidata.org/wiki/Q4117718","display_name":"Program synthesis","level":2,"score":0.17900000512599945},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.1745000034570694},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.17249999940395355},{"id":"https://openalex.org/C134752490","wikidata":"https://www.wikidata.org/wiki/Q374182","display_name":"Logical consequence","level":2,"score":0.1720000058412552},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.1712999939918518},{"id":"https://openalex.org/C162319229","wikidata":"https://www.wikidata.org/wiki/Q175263","display_name":"Data structure","level":2,"score":0.1712000072002411}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2504.14523","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2504.14523","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2504.14523","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2504.14523","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":"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"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education","score":0.4222084581851959}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Training":[0],"models":[1,27,135,206],"on":[2,198,208,220],"synthetic":[3,61,116,210,234],"data":[4,38,62,117,211,235],"has":[5],"emerged":[6],"as":[7],"an":[8,126,221],"increasingly":[9],"important":[10],"strategy":[11],"for":[12,24,51,115,192],"improving":[13,193],"the":[14,31,60,68,78,97,123,155,194,215,229],"performance":[15,195,216],"of":[16,34,46,71,99,125,196,217,224,232],"generative":[17],"AI.":[18],"This":[19],"approach":[20,114,184],"is":[21,120],"particularly":[22],"helpful":[23],"large":[25,53,173],"multimodal":[26,54,174],"(LMMs)":[28],"due":[29],"to":[30,40,63,96,136,153,166,237],"relative":[32],"scarcity":[33],"high-quality":[35],"paired":[36],"image-text":[37],"compared":[39],"language-only":[41],"data.":[42],"While":[43],"a":[44,87,112,142,172],"variety":[45],"methods":[47],"have":[48,103],"been":[49],"proposed":[50],"generating":[52,233],"datasets,":[55],"they":[56,102],"do":[57],"not":[58],"tailor":[59],"address":[64],"specific":[65,238],"deficiencies":[66],"in":[67,86,122,242],"reasoning":[69,100,129,156,239],"abilities":[70],"LMMs":[72,197,218],"which":[73,119,149,161],"will":[74,245],"be":[75,151],"trained":[76,207,219],"with":[77],"generated":[79],"dataset.":[80],"In":[81],"contrast,":[82],"humans":[83],"often":[84],"learn":[85],"more":[88],"efficient":[89],"manner":[90],"by":[91,107,141],"seeking":[92],"out":[93],"examples":[94,148,181],"related":[95],"types":[98],"where":[101],"failed":[104],"previously.":[105],"Inspired":[106],"this":[108],"observation,":[109],"we":[110],"propose":[111,146],"new":[113,147],"generation":[118],"grounded":[121],"analysis":[124],"existing":[127],"LMM's":[128],"failures.":[130],"Our":[131,202],"methodology":[132],"leverages":[133],"frontier":[134],"automatically":[137],"analyze":[138],"errors":[139],"produced":[140],"weaker":[143],"LMM":[144],"and":[145,185,249],"can":[150,212],"used":[152],"correct":[154],"failure":[157,240],"via":[158],"additional":[159,225],"training,":[160],"are":[162],"then":[163],"further":[164],"filtered":[165],"ensure":[167],"high":[168,230],"quality.":[169],"We":[170,244],"generate":[171],"instruction":[175],"tuning":[176],"dataset":[177,248],"containing":[178],"over":[179],"553k":[180],"using":[182],"our":[183,209,247],"conduct":[186],"extensive":[187],"experiments":[188],"demonstrating":[189,228],"its":[190],"utility":[191],"multiple":[199],"downstream":[200],"tasks.":[201],"results":[203],"show":[204],"that":[205],"even":[213],"exceed":[214],"equivalent":[222],"amount":[223],"real":[226],"data,":[227],"value":[231],"targeted":[236],"modes":[241],"LMMs.":[243],"make":[246],"code":[250],"publicly":[251],"available.":[252]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2025-10-10T00:00:00"}