{"id":"https://openalex.org/W7155063525","doi":"https://doi.org/10.1109/tip.2026.3684401","title":"RSTFA: Efficient Training-Free Human-Preference Alignment via Rejection Sampling for Text-to-Image Diffusion Models","display_name":"RSTFA: Efficient Training-Free Human-Preference Alignment via Rejection Sampling for Text-to-Image Diffusion Models","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7155063525","doi":"https://doi.org/10.1109/tip.2026.3684401","pmid":"https://pubmed.ncbi.nlm.nih.gov/42013252"},"language":"en","primary_location":{"id":"doi:10.1109/tip.2026.3684401","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tip.2026.3684401","pdf_url":null,"source":{"id":"https://openalex.org/S4210173141","display_name":"IEEE Transactions on Image Processing","issn_l":"1057-7149","issn":["1057-7149","1941-0042"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["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 Image Processing","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":null,"display_name":"Hongzheng Yang","orcid":"https://orcid.org/0000-0003-2628-5916"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Hongzheng Yang","raw_affiliation_strings":["The Chinese University of Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0000-0003-2628-5916","affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong, Hong Kong, China","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134169810","display_name":"Jason Chun-Lok Li","orcid":"https://orcid.org/0000-0002-9389-4095"},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Jason Chun-Lok Li","raw_affiliation_strings":["University of Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0000-0002-9389-4095","affiliations":[{"raw_affiliation_string":"University of Hong Kong, Hong Kong, China","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085028247","display_name":"K Li","orcid":null},"institutions":[{"id":"https://openalex.org/I168719708","display_name":"City University of Hong Kong","ror":"https://ror.org/03q8dnn23","country_code":"HK","type":"education","lineage":["https://openalex.org/I168719708"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Kun Li","raw_affiliation_strings":["City University of Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0009-0006-2944-3501","affiliations":[{"raw_affiliation_string":"City University of Hong Kong, Hong Kong, China","institution_ids":["https://openalex.org/I168719708"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5019349624","display_name":"Wenao Ma","orcid":"https://orcid.org/0000-0002-9952-5266"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Wenao Ma","raw_affiliation_strings":["The Chinese University of Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0000-0002-9952-5266","affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong, Hong Kong, China","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134153515","display_name":"Mingjie Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mingjie Xu","raw_affiliation_strings":["Hong Kong University of Science and Technology (Guangzhou), Guangzhou, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hong Kong University of Science and Technology (Guangzhou), Guangzhou, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5071193469","display_name":"Yuzhi Zhao","orcid":"https://orcid.org/0000-0001-8561-2206"},"institutions":[{"id":"https://openalex.org/I168719708","display_name":"City University of Hong Kong","ror":"https://ror.org/03q8dnn23","country_code":"HK","type":"education","lineage":["https://openalex.org/I168719708"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Yuzhi Zhao","raw_affiliation_strings":["City University of Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0000-0001-8561-2206","affiliations":[{"raw_affiliation_string":"City University of Hong Kong, Hong Kong, China","institution_ids":["https://openalex.org/I168719708"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5038133707","display_name":"Lai-Man Po","orcid":"https://orcid.org/0000-0002-5185-1492"},"institutions":[{"id":"https://openalex.org/I168719708","display_name":"City University of Hong Kong","ror":"https://ror.org/03q8dnn23","country_code":"HK","type":"education","lineage":["https://openalex.org/I168719708"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Lai-Man Po","raw_affiliation_strings":["City University of Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0000-0002-5185-1492","affiliations":[{"raw_affiliation_string":"City University of Hong Kong, Hong Kong, China","institution_ids":["https://openalex.org/I168719708"]}]}],"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":0.46160183,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"35","issue":null,"first_page":"4571","last_page":"4586"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.43299999833106995,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.43299999833106995,"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":0.08259999752044678,"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/T11448","display_name":"Face recognition and analysis","score":0.05609999969601631,"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/sampling","display_name":"Sampling (signal processing)","score":0.4602000117301941},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3797000050544739},{"id":"https://openalex.org/keywords/diffusion","display_name":"Diffusion","score":0.3456999957561493},{"id":"https://openalex.org/keywords/image-processing","display_name":"Image processing","score":0.2874000072479248},{"id":"https://openalex.org/keywords/estimation-theory","display_name":"Estimation theory","score":0.28200000524520874},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.28040000796318054},{"id":"https://openalex.org/keywords/preference","display_name":"Preference","score":0.2793000042438507},{"id":"https://openalex.org/keywords/noise-measurement","display_name":"Noise measurement","score":0.2745000123977661}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.541100025177002},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.4602000117301941},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4494999945163727},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.44920000433921814},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3797000050544739},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3776000142097473},{"id":"https://openalex.org/C69357855","wikidata":"https://www.wikidata.org/wiki/Q163214","display_name":"Diffusion","level":2,"score":0.3456999957561493},{"id":"https://openalex.org/C9417928","wikidata":"https://www.wikidata.org/wiki/Q1070689","display_name":"Image processing","level":3,"score":0.2874000072479248},{"id":"https://openalex.org/C167928553","wikidata":"https://www.wikidata.org/wiki/Q1376021","display_name":"Estimation theory","level":2,"score":0.28200000524520874},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.28040000796318054},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.2793000042438507},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.2754000127315521},{"id":"https://openalex.org/C29265498","wikidata":"https://www.wikidata.org/wiki/Q7047719","display_name":"Noise measurement","level":3,"score":0.2745000123977661},{"id":"https://openalex.org/C104267543","wikidata":"https://www.wikidata.org/wiki/Q208163","display_name":"Signal processing","level":3,"score":0.27219998836517334},{"id":"https://openalex.org/C2988922011","wikidata":"https://www.wikidata.org/wiki/Q5449244","display_name":"Filtering theory","level":2,"score":0.2721000015735626},{"id":"https://openalex.org/C158457486","wikidata":"https://www.wikidata.org/wiki/Q17104301","display_name":"Multidimensional systems","level":2,"score":0.2653999924659729},{"id":"https://openalex.org/C52740198","wikidata":"https://www.wikidata.org/wiki/Q1539564","display_name":"Importance sampling","level":3,"score":0.2644999921321869},{"id":"https://openalex.org/C8272713","wikidata":"https://www.wikidata.org/wiki/Q176737","display_name":"Stochastic process","level":2,"score":0.26409998536109924},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.2624000012874603},{"id":"https://openalex.org/C187192777","wikidata":"https://www.wikidata.org/wiki/Q381699","display_name":"Rejection sampling","level":5,"score":0.26109999418258667},{"id":"https://openalex.org/C106516650","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm design","level":2,"score":0.25850000977516174},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.2556999921798706}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/tip.2026.3684401","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tip.2026.3684401","pdf_url":null,"source":{"id":"https://openalex.org/S4210173141","display_name":"IEEE Transactions on Image Processing","issn_l":"1057-7149","issn":["1057-7149","1941-0042"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["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 Image Processing","raw_type":"journal-article"},{"id":"pmid:42013252","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/42013252","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 image processing : a publication of the IEEE Signal Processing Society","raw_type":null}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.4853493869304657,"id":"https://metadata.un.org/sdg/10","display_name":"Reduced inequalities"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Given":[0],"a":[1,72,98,103],"text-to-image":[2],"diffusion":[3,32,37,157],"model":[4,14,158],"pretrained":[5],"on":[6,67,126],"large-scale":[7],"text-image":[8],"pairs,":[9],"can":[10],"we":[11,24,70,112],"align":[12],"the":[13,26,36],"with":[15,142],"human":[16,87,143],"pReferences":[17],"without":[18,90],"further":[19],"fine-tuning?":[20],"In":[21],"this":[22,68],"paper,":[23],"analyze":[25],"effect":[27],"of":[28],"alignment":[29,49,64,75,89,109,141],"tuning":[30,50,122],"in":[31],"models":[33],"by":[34],"comparing":[35],"denoising":[38],"trajectory":[39],"between":[40],"base":[41],"and":[42,101,131],"aligned":[43],"models.":[44],"Our":[45],"findings":[46],"reveal":[47],"that":[48,78,114,134],"primarily":[51],"affects":[52],"superficial":[53,63],"stylistic":[54,84],"aspects":[55],"during":[56],"denoising,":[57],"rather":[58],"than":[59,120],"fundamental":[60],"content,":[61],"suggesting":[62],"behaviors.":[65],"Based":[66],"discovery,":[69],"introduce":[71],"novel,":[73],"training-free":[74],"approach":[76],"(RSTFA)":[77],"leverages":[79],"rejection":[80],"sampling":[81],"at":[82],"specific":[83],"timesteps,":[85],"ensuring":[86],"preference":[88],"fine-tuning":[91],"or":[92],"heavy":[93],"inference":[94],"overhead.":[95],"We":[96],"provide":[97],"theoretical":[99],"analysis":[100],"derive":[102],"bias":[104],"bound":[105],"for":[106],"our":[107,135],"rejection-sampling":[108],"scheme.":[110],"Empirically,":[111],"show":[113,133],"RSTFA":[115],"better":[116],"preserves":[117],"sample":[118],"diversity":[119],"reinforcement-learning-based":[121],"methods.":[123],"Extensive":[124],"experiments":[125],"Pick-a-Pic,":[127],"COCO,":[128],"HPD":[129],"V2,":[130],"PartiPrompts":[132],"method":[136],"not":[137],"only":[138],"achieves":[139],"superior":[140],"preferences":[144],"compared":[145],"to":[146],"state-of-the-art":[147],"methods,":[148],"but":[149],"also":[150],"reduces":[151],"computational":[152],"demands,":[153],"establishing":[154],"efficient,":[155],"human-centered":[156],"alignment.":[159]},"counts_by_year":[],"updated_date":"2026-05-07T06:04:25.777469","created_date":"2026-04-22T00:00:00"}