{"id":"https://openalex.org/W7166845561","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1869","title":"PyraMathBench: Evaluating and Improving Mathematical Capability in Large Language Models","display_name":"PyraMathBench: Evaluating and Improving Mathematical Capability in Large Language Models","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166845561","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1869"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.1869","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1869","pdf_url":"https://aclanthology.org/2026.findings-acl.1869.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.findings-acl.1869.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5137788842","display_name":"Zetian Ouyang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zetian Ouyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139760145","display_name":"Linlin Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Linlin Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139734594","display_name":"Gerard de Melo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gerard de Melo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5139741042","display_name":"Liang He","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liang He","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":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.84115659,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"37481","last_page":"37511"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T13523","display_name":"Mathematics, Computing, and Information Processing","score":0.6371999979019165,"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"}},"topics":[{"id":"https://openalex.org/T13523","display_name":"Mathematics, Computing, and Information Processing","score":0.6371999979019165,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.034699998795986176,"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/T14079","display_name":"Mathematics Education and Programs","score":0.025499999523162842,"subfield":{"id":"https://openalex.org/subfields/2611","display_name":"Modeling and Simulation"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/mathematical-model","display_name":"Mathematical model","score":0.30390000343322754},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.2775999903678894},{"id":"https://openalex.org/keywords/quality","display_name":"Quality (philosophy)","score":0.2671999931335449},{"id":"https://openalex.org/keywords/measure","display_name":"Measure (data warehouse)","score":0.2630999982357025},{"id":"https://openalex.org/keywords/identification","display_name":"Identification (biology)","score":0.2614000141620636}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6611999869346619},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4090999960899353},{"id":"https://openalex.org/C76969082","wikidata":"https://www.wikidata.org/wiki/Q486902","display_name":"Mathematical model","level":2,"score":0.30390000343322754},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.2775999903678894},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.2671999931335449},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.2630999982357025},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.2614000141620636},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2556999921798706},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.25529998540878296},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.2513999938964844}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.1869","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1869","pdf_url":"https://aclanthology.org/2026.findings-acl.1869.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-acl.1869","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1869","pdf_url":"https://aclanthology.org/2026.findings-acl.1869.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G4748013168","display_name":null,"funder_award_id":"25JS2830402","funder_id":"https://openalex.org/F4320321885","funder_display_name":"Science and Technology Commission of Shanghai Municipality"},{"id":"https://openalex.org/G4976452659","display_name":null,"funder_award_id":"25JS2830400","funder_id":"https://openalex.org/F4320321885","funder_display_name":"Science and Technology Commission of Shanghai Municipality"},{"id":"https://openalex.org/G673242103","display_name":null,"funder_award_id":"2025SHZDZX025G06","funder_id":"https://openalex.org/F4320321885","funder_display_name":"Science and Technology Commission of Shanghai Municipality"}],"funders":[{"id":"https://openalex.org/F4320321885","display_name":"Science and Technology Commission of Shanghai Municipality","ror":"https://ror.org/03kt66j61"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166845561.pdf","grobid_xml":"https://content.openalex.org/works/W7166845561.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Despite":[0],"the":[1,8,32,88,178],"pivotal":[2],"role":[3],"of":[4,10,34,80,158],"numerical":[5,26,75,82],"reasoning":[6],"as":[7,200],"cornerstone":[9],"mathematical":[11,29],"capabilities":[12],"in":[13,36],"large":[14],"language":[15],"models":[16],"(LLMs)":[17],"across":[18],"applications,":[19],"few":[20],"benchmarks":[21],"evaluate":[22],"LLMs":[23],"by":[24,73],"integrating":[25],"processing":[27],"and":[28,63,77,96,113,127,152],"reasoning,":[30],"hindering":[31],"interpretability":[33],"failures":[35],"math":[37,53],"tasks.We":[38],"introduce":[39],"PyraMath-Bench":[40],"1":[41,132],",":[42],"a":[43,121,156],"comprehensive":[44],"hierarchical":[45],"benchmark":[46],"with":[47,125],"32,505":[48],"questions":[49],"derived":[50],"from":[51,177],"7,404":[52],"word":[54],"problems,":[55],"spanning":[56],"4":[57],"key":[58,150,172],"cognitive":[59],"aspects,":[60],"14":[61],"subcategories,":[62],"2":[64],"modalities.Experiments":[65],"reveal":[66],"that":[67],"LLMs'":[68,104],"performance":[69],"is":[70],"severely":[71],"compromised":[72],"inadequate":[74],"computation":[76],"weak":[78],"handling":[79],"abstract":[81],"questions.To":[83],"address":[84],"this,":[85],"we":[86],"propose":[87],"Smart":[89],"Optimization":[90,100],"&":[91],"Learning-based":[92],"VErsatile":[93],"module":[94],"(SOLVE)":[95],"Interactive":[97],"Relative":[98],"Policy":[99],"(IRPO),":[101],"which":[102],"enhance":[103],"numerical-mathematical":[105],"synergy":[106],"via":[107],"efficient":[108],"tool":[109],"calls":[110],"(fuzzy":[111],"matching":[112],"low-quality":[114],"call":[115],"rejection).Comparative":[116],"experiments":[117],"show":[118],"Qwen-2.5":[119],"achieves":[120],"5.0":[122],"score":[123],"improvement":[124],"SOLVE":[126],"IRPO":[128],"training.*":[129],"Corresponding":[130],"author.":[131],"https://github.com/optifine233-ship-it/Key":[133],"Units:":[134],"30":[135],"(meters^2)":[136],"22":[137],"(meters)":[138],"To":[139,161],"solve":[140,155,162],"this":[141,163],"MWP,":[142],"I":[143,165],"have":[144,166],"to":[145,167],"1.":[146,168],"Abstracting":[147,169],"problems.2.":[148,170],"extract":[149,171,175],"numbers":[151],"units,":[153],"3.":[154,174],"system":[157],"equations,":[159],"\u2026":[160,180],"VRP,":[164],"numbers,":[173],"information":[176],"picture,":[179],"L=6,":[181],"W=5":[182,193],"or":[183],"L=5,":[184],"W=6":[185],"And":[186],"L":[187,206],">":[188],"W,":[189],"so":[190],"L=6":[191],"meters,":[192],"meters":[194],"The":[195],"question":[196],"can":[197],"be":[198],"abstracted":[199],"Unknown":[201],"Variables:":[202],"L,W":[203],"Equations:":[204],"2*(L+W)=22,":[205]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}