{"id":"https://openalex.org/W7224875583","doi":"https://doi.org/10.48550/arxiv.2610.39948","title":"UniRA: Unified Representation Alignment and Auto-Encoding for Efficient Diffusion Generation","display_name":"UniRA: Unified Representation Alignment and Auto-Encoding for Efficient Diffusion Generation","publication_year":2026,"publication_date":"2026-10-29","ids":{"openalex":"https://openalex.org/W7224875583","doi":"https://doi.org/10.48550/arxiv.2610.39948"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.39948","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.39948","pdf_url":"https://arxiv.org/pdf/2610.39948","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"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":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2610.39948","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5146400819","display_name":"An Lanji","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lanji, An","raw_affiliation_strings":["University of Electronic Science and Technology of China , Chengdu , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China , Chengdu , China","institution_ids":["https://openalex.org/I150229711"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5160219199","display_name":"Dawei Liu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Liu, Dawei","raw_affiliation_strings":["University of Electronic Science and Technology of China , Chengdu , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China , Chengdu , China","institution_ids":["https://openalex.org/I150229711"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5160231153","display_name":"Jin Li","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Li, Jin","raw_affiliation_strings":["University of Electronic Science and Technology of China , Chengdu , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China , Chengdu , China","institution_ids":["https://openalex.org/I150229711"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5160221913","display_name":"Haoran Xu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xu, Haoran","raw_affiliation_strings":["University of Electronic Science and Technology of China , Chengdu , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China , Chengdu , China","institution_ids":["https://openalex.org/I150229711"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5160208840","display_name":"Mei Chen","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chen, Mei","raw_affiliation_strings":["University of Electronic Science and Technology of China , Chengdu , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China , Chengdu , China","institution_ids":["https://openalex.org/I150229711"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5160207393","display_name":"Yu Tian","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tian, Yu","raw_affiliation_strings":["University of Electronic Science and Technology of China , Chengdu , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China , Chengdu , China","institution_ids":["https://openalex.org/I150229711"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I150229711"],"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":null,"last_page":null},"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":1.0,"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":1.0,"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":0.0,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.0,"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/diffusion-transformer","display_name":"diffusion transformer","score":0.8040000200271606},{"id":"https://openalex.org/keywords/representation-alignment","display_name":"representation alignment","score":0.8960000276565552},{"id":"https://openalex.org/keywords/image-generation","display_name":"image generation","score":0.6119999885559082},{"id":"https://openalex.org/keywords/training-efficiency","display_name":"training efficiency","score":0.6890000104904175},{"id":"https://openalex.org/keywords/dinov2","display_name":"DINOv2","score":0.718999981880188},{"id":"https://openalex.org/keywords/latent-diffusion-models","display_name":"latent diffusion models","score":0.6190000176429749},{"id":"https://openalex.org/keywords/image-classification","display_name":"image classification","score":0.3619999885559082},{"id":"https://openalex.org/keywords/linear-probing","display_name":"linear probing","score":0.8820000290870667},{"id":"https://openalex.org/keywords/zero-shot-classification","display_name":"zero-shot classification","score":0.7409999966621399},{"id":"https://openalex.org/keywords/joint-optimization","display_name":"joint optimization","score":0.34200000762939453}],"concepts":[{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.7990999817848206},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7179999947547913},{"id":"https://openalex.org/C183115368","wikidata":"https://www.wikidata.org/wiki/Q856577","display_name":"Weighting","level":2,"score":0.5232999920845032},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4507000148296356},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.39070001244544983},{"id":"https://openalex.org/C45493050","wikidata":"https://www.wikidata.org/wiki/Q7884934","display_name":"Unified Model","level":2,"score":0.3774999976158142},{"id":"https://openalex.org/C69357855","wikidata":"https://www.wikidata.org/wiki/Q163214","display_name":"Diffusion","level":2,"score":0.37049999833106995},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.3635999858379364},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.33899998664855957},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.3264999985694885},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.30000001192092896},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2915000021457672},{"id":"https://openalex.org/C68028875","wikidata":"https://www.wikidata.org/wiki/Q7572595","display_name":"Space mapping","level":2,"score":0.27059999108314514}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2610.39948","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.39948","pdf_url":"https://arxiv.org/pdf/2610.39948","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"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":"text"},{"id":"doi:10.48550/arxiv.2610.39948","doi":"https://doi.org/10.48550/arxiv.2610.39948","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2610.39948","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2610.39948","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.39948","pdf_url":"https://arxiv.org/pdf/2610.39948","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"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":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7224875583.pdf","grobid_xml":"https://content.openalex.org/works/W7224875583.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4403345601","https://openalex.org/W7090942304","https://openalex.org/W4414577737","https://openalex.org/W7125800819","https://openalex.org/W6929487296","https://openalex.org/W4416901645","https://openalex.org/W4417511319","https://openalex.org/W7163154248","https://openalex.org/W7212452710","https://openalex.org/W6891798853"],"abstract_inverted_index":{"Diffusion":[0],"transformers":[1],"(DiTs)":[2],"have":[3],"become":[4],"the":[5,56,92,97,127,131,135,137,141,158,162,177,214],"dominant":[6],"architecture":[7],"for":[8,84],"high-fidelity":[9],"image":[10],"synthesis,":[11],"yet":[12],"they":[13],"suffer":[14],"from":[15,30,65,231],"slow":[16],"training":[17,38,204],"convergence":[18],"and":[19,73,87,110,130,140,169,180,189,222,228],"high":[20],"computational":[21],"cost.":[22],"Two":[23],"recent":[24],"lines":[25],"of":[26,161],"work":[27],"address":[28],"this":[29,154],"complementary":[31],"angles:":[32],"Representation":[33,52],"Alignment":[34],"(REPA)":[35],"accelerates":[36,96],"DiT":[37],"by":[39,206],"aligning":[40],"intermediate":[41],"features":[42],"with":[43,59],"a":[44,60,81,103,114,119,149,172,232],"pretrained":[45],"external":[46,142,163],"representation":[47,108,111,143,159,216],"model":[48,144,164],"(e.g.,":[49],"DINOv2),":[50],"while":[51,165,202],"Auto-Encoders":[53],"(RAE)":[54],"replace":[55],"pixel-space":[57,82],"VAE":[58,83],"compact":[61],"representation-space":[62,120],"tokenizer":[63,121],"learned":[64],"self-supervised":[66],"features.":[67],"However,":[68],"these":[69],"approaches":[70],"are":[71,145],"developed":[72],"applied":[74],"in":[75,200],"isolation:":[76],"REPA":[77],"still":[78],"relies":[79],"on":[80],"latent":[85,235],"diffusion,":[86],"RAE":[88],"does":[89],"not":[90],"exploit":[91],"alignment":[93,109,132],"signal":[94],"that":[95,106,153,175,213,226],"diffusion":[98,128,138,167],"backbone.":[99],"We":[100,147,210],"present":[101],"\\textbf{UniRA},":[102],"unified":[104,215],"framework":[105],"couples":[107],"auto-encoding":[112],"into":[113],"single":[115],"objective.":[116],"UniRA":[117,191],"learns":[118],"whose":[122],"output":[123],"serves":[124],"simultaneously":[125],"as":[126],"target":[129],"anchor,":[133],"so":[134],"tokenizer,":[136],"backbone,":[139],"co-optimized.":[146],"give":[148],"theoretical":[150],"result":[151],"showing":[152],"joint":[155],"optimization":[156],"preserves":[157],"quality":[160],"accelerating":[166],"convergence,":[168],"we":[170],"derive":[171],"principled":[173],"weighting":[174],"balances":[176],"reconstruction,":[178],"alignment,":[179],"generation":[181,186,227],"objectives.":[182],"On":[183],"ImageNet":[184],"class-conditional":[185],"at":[187],"256*256":[188],"512*512,":[190],"matches":[192],"or":[193],"surpasses":[194],"state-of-the-art":[195],"DiTs":[196],"(SD3,":[197],"REPA,":[198],"RAE)":[199],"FID":[201],"reducing":[203],"steps":[205],"up":[207],"to":[208],"5*.":[209],"further":[211],"show":[212],"space":[217],"supports":[218],"improved":[219],"zero-shot":[220],"classification":[221],"linear":[223],"probing,":[224],"confirming":[225],"understanding":[229],"benefit":[230],"shared,":[233],"aligned":[234],"space.":[236]},"counts_by_year":[],"updated_date":"2026-11-02T09:45:03.842371","created_date":"2026-10-31T00:00:00"}