{"id":"https://openalex.org/W7163338413","doi":"https://doi.org/10.48550/arxiv.2606.03303","title":"LEAP: Supercharging LLMs for Formal Mathematics with Agentic Frameworks","display_name":"LEAP: Supercharging LLMs for Formal Mathematics with Agentic Frameworks","publication_year":2026,"publication_date":"2026-06-02","ids":{"openalex":"https://openalex.org/W7163338413","doi":"https://doi.org/10.48550/arxiv.2606.03303"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.03303","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.03303","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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.2606.03303","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5137754088","display_name":"Po-Nien Kung","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kung, Po-Nien","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137737224","display_name":"Linfeng Song","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Song, Linfeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5049491751","display_name":"Dawsen Hwang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hwang, Dawsen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137716037","display_name":"Jinsung Yoon","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yoon, Jinsung","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137778701","display_name":"Chun-Liang Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Chun-Liang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5064938440","display_name":"Simone Severini","orcid":"https://orcid.org/0000-0001-7305-6759"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Severini, Simone","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137765222","display_name":"Mirek Ol\u0161\u00e1k","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ol\u0161\u00e1k, Mirek","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5121296619","display_name":"Edward Lockhart","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lockhart, Edward","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137798937","display_name":"Quoc V Le","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Le, Quoc V","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103372430","display_name":"Burak G\u00f6kt\u00fcrk","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gokturk, Burak","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137781430","display_name":"Thang Luong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Luong, Thang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137796271","display_name":"Tomas Pfister","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pfister, Tomas","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5030248499","display_name":"Nanyun Peng","orcid":"https://orcid.org/0000-0002-8509-6595"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Peng, Nanyun","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/T11948","display_name":"Machine Learning in Materials Science","score":0.16349999606609344,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11948","display_name":"Machine Learning in Materials Science","score":0.16349999606609344,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials 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":0.09319999814033508,"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/T10126","display_name":"Logic, programming, and type systems","score":0.07519999891519547,"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/mathematical-proof","display_name":"Mathematical proof","score":0.7908999919891357},{"id":"https://openalex.org/keywords/formal-proof","display_name":"Formal proof","score":0.5012999773025513},{"id":"https://openalex.org/keywords/blueprint","display_name":"Blueprint","score":0.48240000009536743},{"id":"https://openalex.org/keywords/formal-methods","display_name":"Formal methods","score":0.47269999980926514},{"id":"https://openalex.org/keywords/matching","display_name":"Matching (statistics)","score":0.45509999990463257},{"id":"https://openalex.org/keywords/formal-system","display_name":"Formal system","score":0.4072999954223633},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.39149999618530273},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.38839998841285706},{"id":"https://openalex.org/keywords/proof-assistant","display_name":"Proof assistant","score":0.34049999713897705}],"concepts":[{"id":"https://openalex.org/C108710211","wikidata":"https://www.wikidata.org/wiki/Q11538","display_name":"Mathematical proof","level":2,"score":0.7908999919891357},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5131000280380249},{"id":"https://openalex.org/C94461902","wikidata":"https://www.wikidata.org/wiki/Q2762418","display_name":"Formal proof","level":3,"score":0.5012999773025513},{"id":"https://openalex.org/C155911762","wikidata":"https://www.wikidata.org/wiki/Q422321","display_name":"Blueprint","level":2,"score":0.48240000009536743},{"id":"https://openalex.org/C75606506","wikidata":"https://www.wikidata.org/wiki/Q1049183","display_name":"Formal methods","level":2,"score":0.47269999980926514},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.45509999990463257},{"id":"https://openalex.org/C102315432","wikidata":"https://www.wikidata.org/wiki/Q649732","display_name":"Formal system","level":2,"score":0.4072999954223633},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.39149999618530273},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.38839998841285706},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3456000089645386},{"id":"https://openalex.org/C203265346","wikidata":"https://www.wikidata.org/wiki/Q11387554","display_name":"Proof assistant","level":3,"score":0.34049999713897705},{"id":"https://openalex.org/C47884741","wikidata":"https://www.wikidata.org/wiki/Q1166618","display_name":"Mathematical logic","level":2,"score":0.32199999690055847},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.32190001010894775},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.3124000132083893},{"id":"https://openalex.org/C145420912","wikidata":"https://www.wikidata.org/wiki/Q853077","display_name":"Mathematics education","level":1,"score":0.3077999949455261},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.30720001459121704},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.30059999227523804},{"id":"https://openalex.org/C14331664","wikidata":"https://www.wikidata.org/wiki/Q3417382","display_name":"Refinement","level":3,"score":0.2955999970436096},{"id":"https://openalex.org/C539667460","wikidata":"https://www.wikidata.org/wiki/Q2414942","display_name":"Management science","level":1,"score":0.28839999437332153},{"id":"https://openalex.org/C168773036","wikidata":"https://www.wikidata.org/wiki/Q264164","display_name":"Recursion (computer science)","level":2,"score":0.2824000120162964},{"id":"https://openalex.org/C116253237","wikidata":"https://www.wikidata.org/wiki/Q1437424","display_name":"Formal specification","level":2,"score":0.2766000032424927},{"id":"https://openalex.org/C2779231336","wikidata":"https://www.wikidata.org/wiki/Q7534724","display_name":"Sketch","level":2,"score":0.2711000144481659},{"id":"https://openalex.org/C146072743","wikidata":"https://www.wikidata.org/wiki/Q192161","display_name":"Formal language","level":2,"score":0.27079999446868896},{"id":"https://openalex.org/C2779982483","wikidata":"https://www.wikidata.org/wiki/Q6094420","display_name":"Iterative refinement","level":2,"score":0.2687000036239624},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.26269999146461487},{"id":"https://openalex.org/C2780966255","wikidata":"https://www.wikidata.org/wiki/Q5474306","display_name":"Foundation (evidence)","level":2,"score":0.2590000033378601},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.25870001316070557},{"id":"https://openalex.org/C91306197","wikidata":"https://www.wikidata.org/wiki/Q45767","display_name":"Competition (biology)","level":2,"score":0.2547000050544739}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.03303","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.03303","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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.2606.03303","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.03303","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":[{"score":0.7164508700370789,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"Language":[1],"Models":[2],"(LLMs)":[3],"exhibit":[4],"strong":[5],"informal":[6,48,69],"mathematical":[7,142],"reasoning":[8],"but":[9],"struggle":[10],"to":[11,32,159],"generate":[12],"mechanically":[13],"verifiable":[14],"proofs":[15,106,183],"in":[16,96,128,196],"formal":[17,38,65,141,150],"languages":[18],"like":[19],"Lean.":[20],"We":[21],"present":[22],"LEAP,":[23],"an":[24,121],"agentic":[25],"framework":[26],"that":[27],"enables":[28],"general-purpose":[29,154],"foundation":[30,43],"models":[31],"achieve":[33],"state-of-the-art":[34],"performance":[35],"on":[36,115],"automated":[37],"theorem":[39],"proving.":[40],"LEAP":[41,131,146],"leverages":[42],"model":[44],"capabilities,":[45],"such":[46],"as":[47],"reasoning,":[49],"instruction":[50],"following,":[51],"and":[52,104],"iterative":[53],"self-refinement.":[54],"By":[55],"decomposing":[56],"complex":[57,182],"problems":[58,94],"into":[59],"smaller":[60],"units,":[61],"the":[62,75,116,148,163],"system":[63],"bridges":[64],"proof":[66,191],"construction":[67],"with":[68,74,98],"blueprints":[70],"through":[71],"continuous":[72],"interaction":[73],"Lean":[76],"compiler.":[77],"To":[78],"provide":[79],"a":[80,90,108,168,189,193],"rigorous":[81],"evaluation":[82],"beyond":[83],"increasingly":[84],"saturated":[85],"benchmarks,":[86],"we":[87,174],"introduce":[88],"Lean-IMO-Bench,":[89,145],"benchmark":[91,165],"of":[92,111,153,200],"IMO-style":[93],"formalized":[95],"Lean,":[97],"short":[99],"statements":[100],"yet":[101],"highly":[102],"non-routine":[103],"multi-step":[105],"across":[107],"wide":[109],"range":[110],"difficulty":[112],"levels.":[113],"Empirically,":[114],"latest":[117],"2025":[118],"Putnam":[119],"Competition,":[120],"annual":[122],"mathematics":[123],"competition":[124],"for":[125,184,192],"undergraduate":[126],"students":[127],"North":[129],"America,":[130],"solves":[132],"all":[133],"12":[134],"problems,":[135],"matching":[136],"recent":[137],"breakthroughs":[138],"by":[139,167,179],"frontier":[140],"models.":[143],"On":[144],"boosts":[147],"one-shot":[149],"solve":[151],"rate":[152],"LLMs":[155],"from":[156],"below":[157],"10%":[158],"70%,":[160],"notably":[161],"surpassing":[162],"48%":[164],"set":[166],"specialized,":[169],"gold-medal-caliber":[170],"IMO":[171],"system.":[172],"Furthermore,":[173],"demonstrate":[175],"LEAP's":[176],"research-level":[177],"utility":[178],"autonomously":[180],"formalizing":[181],"open":[185],"combinatorial":[186],"challenges,":[187],"including":[188],"verified":[190],"key":[194],"subproblem":[195],"Knuth's":[197],"Hamiltonian":[198],"decomposition":[199],"even-order":[201],"Cayley":[202],"graphs.":[203]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-04T00:00:00"}