{"id":"https://openalex.org/W7128605167","doi":"https://doi.org/10.1109/tpami.2026.3663759","title":"Enhancing MMDiT-Based Text-to-Image Models for Similar Subject Generation","display_name":"Enhancing MMDiT-Based Text-to-Image Models for Similar Subject Generation","publication_year":2026,"publication_date":"2026-02-11","ids":{"openalex":"https://openalex.org/W7128605167","doi":"https://doi.org/10.1109/tpami.2026.3663759","pmid":"https://pubmed.ncbi.nlm.nih.gov/41671119"},"language":"en","primary_location":{"id":"doi:10.1109/tpami.2026.3663759","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2026.3663759","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/A5006214571","display_name":"Tianyi Wei","orcid":"https://orcid.org/0000-0002-7976-8439"},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Tianyi Wei","raw_affiliation_strings":["S-Lab, Nanyang Technological University, Singapore"],"raw_orcid":"https://orcid.org/0000-0002-7976-8439","affiliations":[{"raw_affiliation_string":"S-Lab, Nanyang Technological University, Singapore","institution_ids":["https://openalex.org/I172675005"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125630379","display_name":"Dongdong Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Dongdong Chen","raw_affiliation_strings":["Microsoft GenAI, Redmond, WA, USA"],"raw_orcid":"https://orcid.org/0000-0002-4642-4373","affiliations":[{"raw_affiliation_string":"Microsoft GenAI, Redmond, WA, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125633986","display_name":"Yifan Zhou","orcid":null},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Yifan Zhou","raw_affiliation_strings":["S-Lab, Nanyang Technological University, Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"S-Lab, Nanyang Technological University, Singapore","institution_ids":["https://openalex.org/I172675005"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5052549072","display_name":"Xingang Pan","orcid":"https://orcid.org/0000-0002-5825-9467"},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Xingang Pan","raw_affiliation_strings":["S-Lab, Nanyang Technological University, Singapore"],"raw_orcid":"https://orcid.org/0000-0002-5825-9467","affiliations":[{"raw_affiliation_string":"S-Lab, Nanyang Technological University, Singapore","institution_ids":["https://openalex.org/I172675005"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"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":0.16202073,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"48","issue":"7","first_page":"7692","last_page":"7706"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.6308000087738037,"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"}},"topics":[{"id":"https://openalex.org/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.6308000087738037,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.2955999970436096,"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/T10028","display_name":"Topic Modeling","score":0.009100000374019146,"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/ambiguity","display_name":"Ambiguity","score":0.5590000152587891},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.5458999872207642},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.5095999836921692},{"id":"https://openalex.org/keywords/benchmarking","display_name":"Benchmarking","score":0.41780000925064087},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.4142000079154968},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.3352000117301941},{"id":"https://openalex.org/keywords/block","display_name":"Block (permutation group theory)","score":0.3287000060081482},{"id":"https://openalex.org/keywords/quality","display_name":"Quality (philosophy)","score":0.31839999556541443}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7930999994277954},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5860999822616577},{"id":"https://openalex.org/C2780522230","wikidata":"https://www.wikidata.org/wiki/Q1140419","display_name":"Ambiguity","level":2,"score":0.5590000152587891},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.5458999872207642},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.5095999836921692},{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.41780000925064087},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.4142000079154968},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.38589999079704285},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.3352000117301941},{"id":"https://openalex.org/C2777210771","wikidata":"https://www.wikidata.org/wiki/Q4927124","display_name":"Block (permutation group theory)","level":2,"score":0.3287000060081482},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.31839999556541443},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.2957000136375427},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.29429998993873596},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.2816999852657318},{"id":"https://openalex.org/C2988416141","wikidata":"https://www.wikidata.org/wiki/Q6031139","display_name":"Information loss","level":2,"score":0.2809999883174896},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.27399998903274536},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.2734000086784363},{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.2624000012874603},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.2563999891281128},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.2540000081062317},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2502000033855438}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/tpami.2026.3663759","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2026.3663759","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:41671119","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/41671119","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},{"id":"pmh:oai:dr.ntu.edu.sg:10356/212236","is_oa":false,"landing_page_url":"https://hdl.handle.net/10356/212236","pdf_url":null,"source":{"id":"https://openalex.org/S4306402609","display_name":"DR-NTU (Nanyang Technological University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I172675005","host_organization_name":"Nanyang Technological University","host_organization_lineage":["https://openalex.org/I172675005"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Journal Article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Representing":[0],"the":[1,7,35,53,76,125,148,162,201],"cutting-edge":[2],"technique":[3],"of":[4,42,158,164,200,204],"text-to-image":[5,189],"models,":[6],"latest":[8],"Multimodal":[9],"Diffusion":[10],"Transformer":[11],"(MMDiT)":[12],"largely":[13],"mitigates":[14],"many":[15],"generation":[16,169],"issues":[17],"existing":[18,177],"in":[19],"previous":[20],"models.":[21],"However,":[22],"we":[23,72,89,113,136],"discover":[24],"that":[25,56,115],"it":[26],"still":[27],"suffers":[28],"from":[29],"subject":[30],"neglect":[31],"or":[32,45],"mixing":[33],"when":[34,119],"input":[36],"text":[37],"prompt":[38],"contains":[39],"multiple":[40,121,186],"subjects":[41,160],"similar":[43,122,159],"semantics":[44],"appearance.":[46],"We":[47],"identify":[48],"three":[49,91],"possible":[50],"ambiguities":[51],"within":[52],"MMDiT":[54,187],"architecture":[55],"cause":[57],"this":[58],"problem:":[59],"Inter-block":[60],"Ambiguity,":[61,64],"Text":[62,97],"Encoder":[63,98],"and":[65,101,142,171,181,195],"Semantic":[66],"Ambiguity.":[67],"To":[68],"address":[69],"these":[70,108],"issues,":[71],"propose":[73,138],"to":[74,106,146],"repair":[75],"ambiguous":[77],"latent":[78],"on-the-fly":[79],"by":[80,128],"test-time":[81],"optimization":[82],"at":[83],"early":[84],"denoising":[85],"steps.":[86],"In":[87],"detail,":[88],"design":[90],"loss":[92,130],"functions:":[93],"Block":[94],"Alignment":[95,99],"Loss,":[96,100,103],"Overlap":[102,139],"each":[104],"tailored":[105],"mitigate":[107],"ambiguities.":[109],"Despite":[110],"significant":[111],"improvements,":[112],"observe":[114],"semantic":[116],"ambiguity":[117],"persists":[118],"generating":[120],"subjects,":[123],"as":[124,192],"guidance":[126],"provided":[127],"overlap":[129],"is":[131],"not":[132],"explicit":[133],"enough.":[134],"Therefore,":[135],"further":[137],"Online":[140],"Detection":[141],"Back-to-Start":[143],"Sampling":[144],"Strategy":[145],"alleviate":[147],"problem.":[149],"Experimental":[150],"results":[151],"on":[152],"a":[153],"newly":[154],"constructed":[155],"challenging":[156],"dataset":[157],"validate":[161],"effectiveness":[163],"our":[165,205],"approach,":[166],"showing":[167],"superior":[168],"quality":[170],"much":[172],"higher":[173],"success":[174],"rates":[175],"over":[176],"methods.":[178],"The":[179],"consistent":[180],"substantial":[182],"improvements":[183],"observed":[184],"across":[185],"based":[188],"models":[190],"such":[191],"SD3,":[193],"SD3.5":[194],"FLUX":[196],"provide":[197],"strong":[198],"evidence":[199],"general":[202],"applicability":[203],"method.":[206]},"counts_by_year":[],"updated_date":"2026-07-29T09:40:50.615796","created_date":"2026-02-12T00:00:00"}