{"id":"https://openalex.org/W7117128592","doi":"https://doi.org/10.1109/tpami.2025.3647880","title":"Toward Unified Expertise: Learning a Single Vision Model From Diverse Perception","display_name":"Toward Unified Expertise: Learning a Single Vision Model From Diverse Perception","publication_year":2025,"publication_date":"2025-12-24","ids":{"openalex":"https://openalex.org/W7117128592","doi":"https://doi.org/10.1109/tpami.2025.3647880","pmid":"https://pubmed.ncbi.nlm.nih.gov/41442297"},"language":"en","primary_location":{"id":"doi:10.1109/tpami.2025.3647880","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2025.3647880","pdf_url":null,"source":{"id":"https://openalex.org/S199944782","display_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","issn_l":"0162-8828","issn":["0162-8828","1939-3539","2160-9292"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","pubmed"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5088638313","display_name":"Z. Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I177605424","display_name":"Amherst College","ror":"https://ror.org/028vqfs63","country_code":"US","type":"education","lineage":["https://openalex.org/I177605424"]},{"id":"https://openalex.org/I24603500","display_name":"University of Massachusetts Amherst","ror":"https://ror.org/0072zz521","country_code":"US","type":"education","lineage":["https://openalex.org/I24603500"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zitian Chen","raw_affiliation_strings":["College of Information and Computer Sciences, University of Massachusetts Amherst, Amherst, MA, USA"],"raw_orcid":"https://orcid.org/0000-0002-9424-4617","affiliations":[{"raw_affiliation_string":"College of Information and Computer Sciences, University of Massachusetts Amherst, Amherst, MA, USA","institution_ids":["https://openalex.org/I177605424","https://openalex.org/I24603500"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022382771","display_name":"Mingyu Ding","orcid":"https://orcid.org/0000-0001-6556-8359"},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Mingyu Ding","raw_affiliation_strings":["Department of Electrical and Computer Engineering, University of California Berkeley, Berkeley, CA, USA"],"raw_orcid":"https://orcid.org/0000-0001-6556-8359","affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, University of California Berkeley, Berkeley, CA, USA","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000201589","display_name":"Y. Shen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yikang Shen","raw_affiliation_strings":["IBM, Cambridge, MA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM, Cambridge, MA, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5045674062","display_name":"Erik Learned-Miller","orcid":"https://orcid.org/0000-0002-3778-9135"},"institutions":[{"id":"https://openalex.org/I177605424","display_name":"Amherst College","ror":"https://ror.org/028vqfs63","country_code":"US","type":"education","lineage":["https://openalex.org/I177605424"]},{"id":"https://openalex.org/I24603500","display_name":"University of Massachusetts Amherst","ror":"https://ror.org/0072zz521","country_code":"US","type":"education","lineage":["https://openalex.org/I24603500"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Erik Learned-Miller","raw_affiliation_strings":["College of Information and Computer Sciences, University of Massachusetts Amherst, Amherst, MA, USA"],"raw_orcid":"https://orcid.org/0000-0002-3778-9135","affiliations":[{"raw_affiliation_string":"College of Information and Computer Sciences, University of Massachusetts Amherst, Amherst, MA, USA","institution_ids":["https://openalex.org/I177605424","https://openalex.org/I24603500"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5040877128","display_name":"Chuang Gan","orcid":"https://orcid.org/0000-0003-4031-5886"},"institutions":[{"id":"https://openalex.org/I177605424","display_name":"Amherst College","ror":"https://ror.org/028vqfs63","country_code":"US","type":"education","lineage":["https://openalex.org/I177605424"]},{"id":"https://openalex.org/I24603500","display_name":"University of Massachusetts Amherst","ror":"https://ror.org/0072zz521","country_code":"US","type":"education","lineage":["https://openalex.org/I24603500"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chuang Gan","raw_affiliation_strings":["College of Information and Computer Sciences, University of Massachusetts Amherst, Amherst, MA, USA"],"raw_orcid":"https://orcid.org/0000-0003-4031-5886","affiliations":[{"raw_affiliation_string":"College of Information and Computer Sciences, University of Massachusetts Amherst, Amherst, MA, USA","institution_ids":["https://openalex.org/I177605424","https://openalex.org/I24603500"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.74035325,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"48","issue":"4","first_page":"4896","last_page":"4908"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.8725000023841858,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.8725000023841858,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.019300000742077827,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.013899999670684338,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/modular-design","display_name":"Modular design","score":0.708299994468689},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.5619000196456909},{"id":"https://openalex.org/keywords/modularity","display_name":"Modularity (biology)","score":0.5503000020980835},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.5239999890327454},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.42579999566078186},{"id":"https://openalex.org/keywords/matching","display_name":"Matching (statistics)","score":0.424699991941452},{"id":"https://openalex.org/keywords/adaptation","display_name":"Adaptation (eye)","score":0.42089998722076416},{"id":"https://openalex.org/keywords/discriminative-model","display_name":"Discriminative model","score":0.3921999931335449},{"id":"https://openalex.org/keywords/modular-neural-network","display_name":"Modular neural network","score":0.37599998712539673}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7598999738693237},{"id":"https://openalex.org/C101468663","wikidata":"https://www.wikidata.org/wiki/Q1620158","display_name":"Modular design","level":2,"score":0.708299994468689},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6409000158309937},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.5619000196456909},{"id":"https://openalex.org/C2779478453","wikidata":"https://www.wikidata.org/wiki/Q6889748","display_name":"Modularity (biology)","level":2,"score":0.5503000020980835},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5472999811172485},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.5239999890327454},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.42579999566078186},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.424699991941452},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.42089998722076416},{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.3921999931335449},{"id":"https://openalex.org/C2781121602","wikidata":"https://www.wikidata.org/wiki/Q3504403","display_name":"Modular neural network","level":4,"score":0.37599998712539673},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.3709999918937683},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.36890000104904175},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.3555999994277954},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.3310999870300293},{"id":"https://openalex.org/C28006648","wikidata":"https://www.wikidata.org/wiki/Q6934509","display_name":"Multi-task learning","level":3,"score":0.32010000944137573},{"id":"https://openalex.org/C94124525","wikidata":"https://www.wikidata.org/wiki/Q912550","display_name":"Categorization","level":2,"score":0.3188999891281128},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.31310001015663147},{"id":"https://openalex.org/C66024118","wikidata":"https://www.wikidata.org/wiki/Q1122506","display_name":"Computational model","level":2,"score":0.31139999628067017},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.30959999561309814},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.2987000048160553},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.29159998893737793},{"id":"https://openalex.org/C152139883","wikidata":"https://www.wikidata.org/wiki/Q252973","display_name":"Mutual information","level":2,"score":0.2818000018596649},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.28049999475479126},{"id":"https://openalex.org/C202615002","wikidata":"https://www.wikidata.org/wiki/Q783507","display_name":"Differentiable function","level":2,"score":0.2736999988555908},{"id":"https://openalex.org/C127162648","wikidata":"https://www.wikidata.org/wiki/Q16858953","display_name":"Channel (broadcasting)","level":2,"score":0.27129998803138733}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/tpami.2025.3647880","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2025.3647880","pdf_url":null,"source":{"id":"https://openalex.org/S199944782","display_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","issn_l":"0162-8828","issn":["0162-8828","1939-3539","2160-9292"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","raw_type":"journal-article"},{"id":"pmid:41442297","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/41442297","pdf_url":null,"source":{"id":"https://openalex.org/S4306525036","display_name":"PubMed","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1299303238","host_organization_name":"National Institutes of Health","host_organization_lineage":["https://openalex.org/I1299303238"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE transactions on pattern analysis and machine intelligence","raw_type":null}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Partnerships for the goals","id":"https://metadata.un.org/sdg/17","score":0.42178937792778015}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":52,"referenced_works":["https://openalex.org/W125693051","https://openalex.org/W1861492603","https://openalex.org/W1977295328","https://openalex.org/W2031489346","https://openalex.org/W2108598243","https://openalex.org/W2125215748","https://openalex.org/W2138011018","https://openalex.org/W2150884987","https://openalex.org/W2340897893","https://openalex.org/W2732026016","https://openalex.org/W2737258237","https://openalex.org/W2797977484","https://openalex.org/W2808168148","https://openalex.org/W2884751099","https://openalex.org/W2884822772","https://openalex.org/W2913340405","https://openalex.org/W2927589347","https://openalex.org/W2962864421","https://openalex.org/W2962935523","https://openalex.org/W2963351448","https://openalex.org/W2963393494","https://openalex.org/W2963430933","https://openalex.org/W2963498646","https://openalex.org/W2963877604","https://openalex.org/W2964185501","https://openalex.org/W2964247799","https://openalex.org/W2981468122","https://openalex.org/W2984039063","https://openalex.org/W2989808579","https://openalex.org/W3035071066","https://openalex.org/W3093412994","https://openalex.org/W3108652056","https://openalex.org/W3138516171","https://openalex.org/W3173563887","https://openalex.org/W3176828726","https://openalex.org/W3204397973","https://openalex.org/W3204647170","https://openalex.org/W4211051725","https://openalex.org/W4221167922","https://openalex.org/W4312231081","https://openalex.org/W4312446817","https://openalex.org/W4312881242","https://openalex.org/W4312884055","https://openalex.org/W4313166855","https://openalex.org/W4385573095","https://openalex.org/W4386076670","https://openalex.org/W4386790226","https://openalex.org/W4388976024","https://openalex.org/W4390872608","https://openalex.org/W4394625552","https://openalex.org/W4402671950","https://openalex.org/W4402727760"],"related_works":[],"abstract_inverted_index":{"Multi-task":[0],"learning":[1,9,103,127],"(MTL)":[2],"presents":[3],"greater":[4],"optimization":[5],"challenges":[6],"than":[7],"single-task":[8,106],"(STL)":[10],"due":[11],"to":[12,97,158,197],"conflicting":[13],"gradients":[14],"across":[15,104,128],"tasks.":[16],"While":[17],"parameter":[18,181],"sharing":[19,78],"promotes":[20],"cooperation":[21,32],"among":[22],"related":[23],"tasks,":[24],"many":[25],"tasks":[26,86,160],"require":[27],"specialized":[28],"representations.":[29],"To":[30],"balance":[31],"and":[33,79,87,161,183,219,231],"specialization,":[34],"we":[35,94,152,187],"propose":[36],"Mod-Squad":[37,96,204],"(Chen":[38],"et":[39],"al.":[40],"2023),":[41],"a":[42,48,55,61,67,117,163,189,206],"modular":[43,73,211],"transformer-based":[44],"model":[45,179],"composed":[46],"of":[47,50,58,85,120,165,178],"\"squad\"":[49],"experts.":[51],"Each":[52],"task":[53],"activates":[54],"sparse":[56,210],"subset":[57],"experts":[59],"through":[60],"differentiable":[62],"matching":[63],"process,":[64],"guided":[65],"by":[66],"novel":[68],"mutual":[69,122],"information-based":[70],"loss.":[71],"This":[72,113],"structure":[74],"avoids":[75],"full":[76],"backbone":[77],"scales":[80],"effectively":[81],"with":[82],"the":[83,121,155],"number":[84],"dataset":[88],"size.":[89],"In":[90,149,202],"this":[91,150],"extended":[92],"version,":[93],"generalize":[95],"support":[98],"multi-dataset":[99],"pre-training,":[100],"enabling":[101],"joint":[102],"disjoint,":[105],"datasets":[107],"(e.g.,":[108],"ImageNet,":[109],"COCO,":[110],"ADE20":[111],"K).":[112],"is":[114],"achieved":[115],"via":[116],"new":[118],"formulation":[119],"information":[123],"loss":[124],"that":[125,169,193,213],"unifies":[126],"heterogeneous":[129],"sources.":[130],"More":[131],"importantly,":[132],"while":[133],"most":[134],"prior":[135],"work":[136],"in":[137],"large":[138],"models":[139,212],"has":[140],"focused":[141],"on":[142],"efficiency,":[143],"few":[144],"have":[145],"explored":[146],"adjustable":[147],"efficiency.":[148],"study,":[151],"further":[153],"evaluate":[154],"model's":[156],"generalization":[157],"downstream":[159,236],"introduce":[162],"set":[164],"efficient":[166],"adaptation":[167,191,234],"techniques":[168,196],"leverage":[170],"Mod-Squad's":[171],"modularity":[172,223],"for":[173,209,235],"flexible":[174],"fine-tuning-enabling":[175],"dynamic":[176],"adjustment":[177],"size,":[180],"count,":[182],"computational":[184],"cost.":[185],"Additionally,":[186],"present":[188],"hybrid":[190],"scheme":[192],"combines":[194],"these":[195],"achieve":[198],"favorable":[199],"performance-efficiency":[200],"trade-offs.":[201],"summary,":[203],"provides":[205],"robust":[207],"foundation":[208],"can":[214],"learn":[215],"from":[216],"diverse":[217],"supervision":[218],"datasets.":[220],"Its":[221],"emergent":[222],"enables":[224],"strong":[225],"generalization,":[226],"decomposition":[227],"into":[228],"high-performing":[229],"components,":[230],"rapid,":[232],"resource-efficient":[233],"applications.":[237]},"counts_by_year":[],"updated_date":"2026-03-09T07:00:12.390032","created_date":"2025-12-24T00:00:00"}