{"id":"https://openalex.org/W7138103432","doi":"https://doi.org/10.1609/aaai.v40i10.37763","title":"From Intent to Execution: Multimodal Chain-of-Thought Reinforcement Learning for Precise CAD Code Generation","display_name":"From Intent to Execution: Multimodal Chain-of-Thought Reinforcement Learning for Precise CAD Code Generation","publication_year":2026,"publication_date":"2026-03-14","ids":{"openalex":"https://openalex.org/W7138103432","doi":"https://doi.org/10.1609/aaai.v40i10.37763"},"language":null,"primary_location":{"id":"doi:10.1609/aaai.v40i10.37763","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i10.37763","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://doi.org/10.1609/aaai.v40i10.37763","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5129647542","display_name":"Ke Niu","orcid":null},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Ke Niu","raw_affiliation_strings":["Fudan University"],"affiliations":[{"raw_affiliation_string":"Fudan University","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129683357","display_name":"Haiyang Yu","orcid":null},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Haiyang Yu","raw_affiliation_strings":["Fudan University"],"affiliations":[{"raw_affiliation_string":"Fudan University","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5121604713","display_name":"Zhuofan Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhuofan Chen","raw_affiliation_strings":["Fudan University"],"affiliations":[{"raw_affiliation_string":"Fudan University","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5065258170","display_name":"Mengyang Zhao","orcid":"https://orcid.org/0000-0001-5952-1263"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Mengyang Zhao","raw_affiliation_strings":["Fudan University"],"affiliations":[{"raw_affiliation_string":"Fudan University","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054421452","display_name":"Teng Fu","orcid":"https://orcid.org/0000-0002-2250-2453"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Teng Fu","raw_affiliation_strings":["Fudan University and ByteDance Inc"],"affiliations":[{"raw_affiliation_string":"Fudan University and ByteDance Inc","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129690513","display_name":"Bin Li","orcid":null},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bin Li","raw_affiliation_strings":["Fudan University"],"affiliations":[{"raw_affiliation_string":"Fudan University","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5129739981","display_name":"Xiangyang Xue","orcid":null},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiangyang Xue","raw_affiliation_strings":["Fudan University"],"affiliations":[{"raw_affiliation_string":"Fudan University","institution_ids":["https://openalex.org/I24943067"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5129647542"],"corresponding_institution_ids":["https://openalex.org/I24943067"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.66666667,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"40","issue":"10","first_page":"8160","last_page":"8167"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11159","display_name":"Manufacturing Process and Optimization","score":0.5656999945640564,"subfield":{"id":"https://openalex.org/subfields/2209","display_name":"Industrial and Manufacturing Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11159","display_name":"Manufacturing Process and Optimization","score":0.5656999945640564,"subfield":{"id":"https://openalex.org/subfields/2209","display_name":"Industrial and Manufacturing Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10719","display_name":"3D Shape Modeling and Analysis","score":0.17599999904632568,"subfield":{"id":"https://openalex.org/subfields/2206","display_name":"Computational Mechanics"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10672","display_name":"Design Education and Practice","score":0.022700000554323196,"subfield":{"id":"https://openalex.org/subfields/2210","display_name":"Mechanical 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/executable","display_name":"Executable","score":0.8485999703407288},{"id":"https://openalex.org/keywords/cad","display_name":"CAD","score":0.6299999952316284},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.586899995803833},{"id":"https://openalex.org/keywords/workflow","display_name":"Workflow","score":0.5737000107765198},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.5665000081062317},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.49399998784065247},{"id":"https://openalex.org/keywords/code-generation","display_name":"Code generation","score":0.44179999828338623}],"concepts":[{"id":"https://openalex.org/C160145156","wikidata":"https://www.wikidata.org/wiki/Q778586","display_name":"Executable","level":2,"score":0.8485999703407288},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8314999938011169},{"id":"https://openalex.org/C194789388","wikidata":"https://www.wikidata.org/wiki/Q17855283","display_name":"CAD","level":2,"score":0.6299999952316284},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.586899995803833},{"id":"https://openalex.org/C177212765","wikidata":"https://www.wikidata.org/wiki/Q627335","display_name":"Workflow","level":2,"score":0.5737000107765198},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.5665000081062317},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.49399998784065247},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46549999713897705},{"id":"https://openalex.org/C133162039","wikidata":"https://www.wikidata.org/wiki/Q1061077","display_name":"Code generation","level":3,"score":0.44179999828338623},{"id":"https://openalex.org/C117251300","wikidata":"https://www.wikidata.org/wiki/Q1849855","display_name":"Parametric statistics","level":2,"score":0.43860000371932983},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.4041000008583069},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.3840999901294708},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.36649999022483826},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.35830000042915344},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3569999933242798},{"id":"https://openalex.org/C24574437","wikidata":"https://www.wikidata.org/wiki/Q7135228","display_name":"Parametric model","level":3,"score":0.3237000107765198},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.3084999918937683},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.29339998960494995},{"id":"https://openalex.org/C158154518","wikidata":"https://www.wikidata.org/wiki/Q7310970","display_name":"Relevance (law)","level":2,"score":0.2865999937057495},{"id":"https://openalex.org/C207685749","wikidata":"https://www.wikidata.org/wiki/Q2088941","display_name":"Domain knowledge","level":2,"score":0.2736000120639801},{"id":"https://openalex.org/C92548554","wikidata":"https://www.wikidata.org/wiki/Q2262868","display_name":"Domain model","level":3,"score":0.26170000433921814}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1609/aaai.v40i10.37763","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i10.37763","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1609/aaai.v40i10.37763","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i10.37763","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Computer-Aided":[0],"Design":[1],"(CAD)":[2],"plays":[3],"a":[4,79,165],"vital":[5],"role":[6],"in":[7,25,196],"engineering":[8],"and":[9,19,70,115,126,150,160,177,185,201],"manufacturing,":[10],"yet":[11],"current":[12],"CAD":[13,56,90,171],"workflows":[14],"require":[15],"extensive":[16],"domain":[17],"expertise":[18],"manual":[20],"modeling":[21,91],"effort.":[22],"Recent":[23],"advances":[24],"large":[26],"language":[27,176,180],"models":[28],"(LLMs)":[29],"have":[30],"made":[31],"it":[32],"possible":[33],"to":[34,62,153],"generate":[35],"code":[36,57,92,202],"from":[37],"natural":[38,175],"language,":[39],"opening":[40],"new":[41],"opportunities":[42],"for":[43,65,89,139,145],"automating":[44],"parametric":[45],"3D":[46,187],"modeling.":[47],"However,":[48],"directly":[49],"translating":[50],"human":[51],"design":[52,179],"intent":[53],"into":[54],"executable":[55,182],"remains":[58],"highly":[59],"challenging,":[60],"due":[61],"the":[63],"need":[64],"logical":[66],"reasoning,":[67],"syntactic":[68],"correctness,":[69],"numerical":[71],"precision.":[72],"In":[73],"this":[74],"work,":[75],"we":[76,130,162],"propose":[77],"CAD-RL,":[78],"multimodal":[80],"Chain-of-Thought":[81],"(CoT)":[82],"guided":[83],"reinforcement":[84,102],"learning":[85,103,123],"post":[86,104],"training":[87,105,159],"framework":[88],"generation.":[93],"Our":[94],"method":[95],"combines":[96],"CoT-based":[97],"Cold":[98],"Start":[99],"with":[100,173],"goal-driven":[101],"using":[106],"three":[107,132],"task-specific":[108],"rewards:":[109],"executability":[110,203],"reward,":[111,114],"geometric":[112],"accuracy":[113],"external":[116],"evaluation":[117],"reward.":[118],"To":[119,157],"ensure":[120],"stable":[121],"policy":[122],"under":[124],"sparse":[125],"high-variance":[127],"reward":[128],"conditions,":[129],"introduce":[131],"targeted":[133],"optimization":[134],"strategies:":[135],"Trust":[136],"Region":[137],"Stretch":[138],"improved":[140],"exploration,":[141],"Precision":[142],"Token":[143],"Loss":[144],"enhanced":[146],"dimensions":[147],"parameter":[148],"accuracy,":[149],"Overlong":[151],"Filtering":[152],"reduce":[154],"noisy":[155],"supervision.":[156],"support":[158],"benchmarking,":[161],"release":[163],"ExeCAD,":[164],"noval":[166],"dataset":[167],"comprising":[168],"16,540":[169],"real-world":[170],"examples":[172],"paired":[174],"structured":[178],"descriptions,":[181],"CADQuery":[183],"scripts,":[184],"rendered":[186],"models.":[188],"Experiments":[189],"demonstrate":[190],"that":[191],"CAD-RL":[192],"achieves":[193],"significant":[194],"improvements":[195],"reasoning":[197],"quality,":[198],"output":[199],"precision,":[200],"over":[204],"existing":[205],"VLMs.":[206]},"counts_by_year":[],"updated_date":"2026-03-20T20:47:17.329874","created_date":"2026-03-18T00:00:00"}