{"id":"https://openalex.org/W4416036485","doi":"https://doi.org/10.18653/v1/2025.emnlp-main.660","title":"From A and B to A+B: Can Large Language Models Solve Compositional Math Problems?","display_name":"From A and B to A+B: Can Large Language Models Solve Compositional Math Problems?","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416036485","doi":"https://doi.org/10.18653/v1/2025.emnlp-main.660"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.emnlp-main.660","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.emnlp-main.660","pdf_url":"https://aclanthology.org/2025.emnlp-main.660.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":"Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.emnlp-main.660.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5072898727","display_name":"X. Xiao","orcid":"https://orcid.org/0009-0002-6595-7172"},"institutions":[{"id":"https://openalex.org/I101479585","display_name":"South China Agricultural University","ror":"https://ror.org/05v9jqt67","country_code":"CN","type":"education","lineage":["https://openalex.org/I101479585"]},{"id":"https://openalex.org/I52158045","display_name":"China Agricultural University","ror":"https://ror.org/04v3ywz14","country_code":"CN","type":"education","lineage":["https://openalex.org/I52158045"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Xisheng Xiao","raw_affiliation_strings":["College of Mathematics and Informatics , South China Agricultural University , China"],"affiliations":[{"raw_affiliation_string":"College of Mathematics and Informatics , South China Agricultural University , China","institution_ids":["https://openalex.org/I101479585","https://openalex.org/I52158045"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5108771122","display_name":"Hanzhang Zhao","orcid":"https://orcid.org/0009-0003-8492-6803"},"institutions":[{"id":"https://openalex.org/I101479585","display_name":"South China Agricultural University","ror":"https://ror.org/05v9jqt67","country_code":"CN","type":"education","lineage":["https://openalex.org/I101479585"]},{"id":"https://openalex.org/I52158045","display_name":"China Agricultural University","ror":"https://ror.org/04v3ywz14","country_code":"CN","type":"education","lineage":["https://openalex.org/I52158045"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hanlin Zhao","raw_affiliation_strings":["College of Mathematics and Informatics , South China Agricultural University , China"],"affiliations":[{"raw_affiliation_string":"College of Mathematics and Informatics , South China Agricultural University , China","institution_ids":["https://openalex.org/I101479585","https://openalex.org/I52158045"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5072898727"],"corresponding_institution_ids":["https://openalex.org/I101479585","https://openalex.org/I52158045"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.28546046,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"13068","last_page":"13089"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11948","display_name":"Machine Learning in Materials Science","score":0.14000000059604645,"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":0.14000000059604645,"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/T12090","display_name":"Language and cultural evolution","score":0.08250000327825546,"subfield":{"id":"https://openalex.org/subfields/3316","display_name":"Cultural Studies"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T12808","display_name":"Ferroelectric and Negative Capacitance Devices","score":0.02539999969303608,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/algebra-over-a-field","display_name":"Algebra over a field","score":0.3125},{"id":"https://openalex.org/keywords/calculus","display_name":"Calculus (dental)","score":0.2759999930858612},{"id":"https://openalex.org/keywords/component","display_name":"Component (thermodynamics)","score":0.2718999981880188},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.25209999084472656}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.48420000076293945},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3946000039577484},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.32829999923706055},{"id":"https://openalex.org/C136119220","wikidata":"https://www.wikidata.org/wiki/Q1000660","display_name":"Algebra over a field","level":2,"score":0.3125},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2969000041484833},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2906999886035919},{"id":"https://openalex.org/C2777686260","wikidata":"https://www.wikidata.org/wiki/Q144037","display_name":"Calculus (dental)","level":2,"score":0.2759999930858612},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.2718999981880188},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2606000006198883},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.25209999084472656}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.emnlp-main.660","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.emnlp-main.660","pdf_url":"https://aclanthology.org/2025.emnlp-main.660.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":"Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.emnlp-main.660","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.emnlp-main.660","pdf_url":"https://aclanthology.org/2025.emnlp-main.660.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":"Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416036485.pdf","grobid_xml":"https://content.openalex.org/works/W4416036485.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"language":[1],"models":[2],"(LLMs)":[3],"have":[4],"demonstrated":[5],"strong":[6],"performance":[7,61,106],"in":[8,77,126],"solving":[9],"math":[10,55],"problems,":[11],"and":[12,57,73,143,157,167],"there":[13],"is":[14,70],"growing":[15],"research":[16],"on":[17,37,62,136,159],"evaluating":[18],"their":[19,113],"robustness.Unlike":[20],"previous":[21],"studies":[22],"that":[23,83,89],"create":[24],"problem":[25],"variants":[26],"by":[27],"adding":[28],"perturbations":[29],"to":[30,51,64,124],"a":[31,48,53,97],"single":[32],"problem,":[33,56],"this":[34],"paper":[35],"focuses":[36],"the":[38,59,92,105,110,141,150],"interaction":[39],"between":[40],"problems.Specifically,":[41],"we":[42,116,148],"combine":[43],"two":[44],"original":[45],"problems":[46],"with":[47,121],"logical":[49],"connection":[50],"get":[52],"new":[54],"measure":[58],"LLM's":[60],"it":[63],"evaluate":[65],"its":[66],"compositional":[67],"generalization,":[68],"which":[69],"an":[71,118],"important":[72,146],"essential":[74],"reasoning":[75],"capability":[76],"human":[78],"intelligence.The":[79],"result":[80],"of":[81,100,107,112,152],"experiments":[82,134],"cover":[84],"14":[85],"different":[86],"LLMs":[87],"shows":[88],"even":[90],"when":[91],"mathematical":[93],"essence":[94],"remains":[95],"unchanged,":[96],"simple":[98],"form":[99],"combination":[101],"can":[102],"significantly":[103],"reduce":[104],"LLMs,":[108],"revealing":[109],"limitation":[111],"generalization":[114],"ability.Additionally,":[115],"propose":[117],"automated":[119],"pipeline":[120],"98.2%":[122],"accuracy":[123],"assist":[125],"annotating":[127],"datasets":[128,138],"(1":[129],"manual,":[130],"2":[131],"synthetic).The":[132],"extensive":[133],"conducted":[135],"these":[137],"further":[139],"verify":[140],"conclusion":[142],"obtain":[144],"some":[145],"findings.Finally,":[147],"analyze":[149],"impact":[151],"factors":[153],"such":[154],"as":[155],"difficulty":[156],"length":[158],"LLMs'":[160],"performance,":[161],"offering":[162],"insights":[163],"for":[164],"future":[165],"research.Code":[166]},"counts_by_year":[],"updated_date":"2026-03-07T13:37:22.277990","created_date":"2025-11-08T00:00:00"}