{"id":"https://openalex.org/W4416034204","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.1233","title":"MEXA: Towards General Multimodal Reasoning with Dynamic Multi-Expert Aggregation","display_name":"MEXA: Towards General Multimodal Reasoning with Dynamic Multi-Expert Aggregation","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416034204","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.1233"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.1233","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.1233","pdf_url":"https://aclanthology.org/2025.findings-emnlp.1233.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: EMNLP 2025","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-emnlp.1233.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5059195621","display_name":"Shoubin Yu","orcid":"https://orcid.org/0009-0006-1670-0054"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shoubin Yu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000258972","display_name":"Y. Zhang","orcid":"https://orcid.org/0009-0007-7271-5979"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yue Zhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100348388","display_name":"Ziyang Wang","orcid":"https://orcid.org/0000-0002-2512-1760"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ziyang Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054456399","display_name":"Jaehong Yoon","orcid":"https://orcid.org/0000-0002-9653-9590"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jaehong Yoon","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5001987532","display_name":"Mohit Bansal","orcid":"https://orcid.org/0000-0001-5522-1351"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mohit Bansal","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":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"22636","last_page":"22652"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12031","display_name":"Speech and dialogue systems","score":0.2628999948501587,"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"}},"topics":[{"id":"https://openalex.org/T12031","display_name":"Speech and dialogue systems","score":0.2628999948501587,"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/T11596","display_name":"Constraint Satisfaction and Optimization","score":0.13300000131130219,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T10906","display_name":"AI-based Problem Solving and Planning","score":0.10769999772310257,"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/feature","display_name":"Feature (linguistics)","score":0.29010000824928284},{"id":"https://openalex.org/keywords/field","display_name":"Field (mathematics)","score":0.2662999927997589},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.25209999084472656},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.24660000205039978},{"id":"https://openalex.org/keywords/non-monotonic-logic","display_name":"Non-monotonic logic","score":0.23199999332427979}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6177999973297119},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5277000069618225},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.29010000824928284},{"id":"https://openalex.org/C9652623","wikidata":"https://www.wikidata.org/wiki/Q190109","display_name":"Field (mathematics)","level":2,"score":0.2662999927997589},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.25209999084472656},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.24660000205039978},{"id":"https://openalex.org/C159032336","wikidata":"https://www.wikidata.org/wiki/Q2488768","display_name":"Non-monotonic logic","level":2,"score":0.23199999332427979},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.23170000314712524},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.23010000586509705},{"id":"https://openalex.org/C20162079","wikidata":"https://www.wikidata.org/wiki/Q1151406","display_name":"Case-based reasoning","level":2,"score":0.22220000624656677}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-emnlp.1233","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.1233","pdf_url":"https://aclanthology.org/2025.findings-emnlp.1233.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: EMNLP 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.1233","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.1233","pdf_url":"https://aclanthology.org/2025.findings-emnlp.1233.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: EMNLP 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G2921176242","display_name":null,"funder_award_id":"W911NF2110220","funder_id":"https://openalex.org/F4320338281","funder_display_name":"Army Research Office"},{"id":"https://openalex.org/G527861410","display_name":null,"funder_award_id":"DRL211263","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G5599941371","display_name":null,"funder_award_id":"N00014-23-1-2356","funder_id":"https://openalex.org/F4320337345","funder_display_name":"Office of Naval Research"},{"id":"https://openalex.org/G8876996369","display_name":null,"funder_award_id":"N00014","funder_id":"https://openalex.org/F4320337345","funder_display_name":"Office of Naval Research"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320332161","display_name":"National Institutes of Health","ror":"https://ror.org/01cwqze88"},{"id":"https://openalex.org/F4320332180","display_name":"Defense Advanced Research Projects Agency","ror":"https://ror.org/02caytj08"},{"id":"https://openalex.org/F4320337345","display_name":"Office of Naval Research","ror":"https://ror.org/00rk2pe57"},{"id":"https://openalex.org/F4320338281","display_name":"Army Research Office","ror":"https://ror.org/05epdh915"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416034204.pdf","grobid_xml":"https://content.openalex.org/works/W4416034204.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Combining":[0],"pre-trained":[1],"expert":[2,67,81,95],"models":[3,68,82],"offers":[4],"substantial":[5],"potential":[6],"for":[7],"scalable":[8],"multimodal":[9,72,131,146,164,180],"reasoning,":[10],"but":[11],"building":[12],"a":[13,57,99,115],"unified":[14],"framework":[15,59],"remains":[16],"challenging":[17],"due":[18],"to":[19,47,69,120],"the":[20,85,89,122,167],"increasing":[21],"diversity":[22],"of":[23,65,172],"input":[24,86],"modalities":[25],"and":[26,76,88,109,129,155,169,176],"task":[27],"complexity.For":[28],"instance,":[29],"medical":[30],"diagnosis":[31],"requires":[32],"precise":[33],"reasoning":[34,73,91,105,132,181],"over":[35,111,162],"structured":[36],"clinical":[37],"tables,":[38],"while":[39],"financial":[40],"forecasting":[41],"depends":[42],"on":[43,84,144],"interpreting":[44],"plot-based":[45],"data":[46],"make":[48],"informed":[49],"predictions.To":[50],"tackle":[51],"this":[52],"challenge,":[53],"we":[54],"introduce":[55],"MEXA,":[56],"training-free":[58],"that":[60],"performs":[61],"modality-and":[62],"task-aware":[63],"aggregation":[64,177],"multiple":[66],"enable":[70],"effective":[71],"across":[74,133],"diverse":[75,134,145,179],"distinct":[77],"domains.MEXA":[78],"dynamically":[79],"selects":[80],"based":[83],"modality":[87],"task-specific":[90],"demands":[92],"(i.e.,":[93],"skills).Each":[94],"model,":[96],"specialized":[97],"in":[98,178],"modalitytask":[100],"pair,":[101],"generates":[102],"interpretable":[103],"textual":[104],"outputs.MEXA":[106],"then":[107],"aggregates":[108],"reasons":[110],"these":[112],"outputs":[113],"using":[114],"Large":[116],"Reasoning":[117],"Model":[118],"(LRM)":[119],"produce":[121],"final":[123],"answer.This":[124],"modular":[125],"design":[126],"allows":[127],"flexible":[128],"transparent":[130],"domains":[135],"without":[136],"additional":[137],"training":[138],"overhead.We":[139],"extensively":[140],"evaluate":[141],"our":[142,173],"approach":[143],"benchmarks,":[147],"including":[148],"Video":[149],"Reasoning,":[150,152],"Audio":[151],"3D":[153],"Understanding,":[154],"Medical":[156],"QA.MEXA":[157],"consistently":[158],"delivers":[159],"performance":[160],"improvements":[161],"strong":[163],"baselines,":[165],"highlighting":[166],"effectiveness":[168],"broad":[170],"applicability":[171],"expert-driven":[174],"selection":[175],"tasks.":[182]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-11-08T00:00:00"}