{"id":"https://openalex.org/W4408353146","doi":"https://doi.org/10.1109/icassp49660.2025.10890823","title":"Positive Enhanced Preference Alignment for Text-to-Image Models","display_name":"Positive Enhanced Preference Alignment for Text-to-Image Models","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408353146","doi":"https://doi.org/10.1109/icassp49660.2025.10890823"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10890823","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10890823","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"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/A5100538111","display_name":"Haoyuan Sun","orcid":"https://orcid.org/0000-0003-1719-8976"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Haoyuan Sun","raw_affiliation_strings":["Tsinghua University,Shenzhen International Graduate School,Shenzhen,China"],"affiliations":[{"raw_affiliation_string":"Tsinghua University,Shenzhen International Graduate School,Shenzhen,China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5044226655","display_name":"Bo Xia","orcid":"https://orcid.org/0000-0001-7694-4743"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bo Xia","raw_affiliation_strings":["Tsinghua University,Shenzhen International Graduate School,Shenzhen,China"],"affiliations":[{"raw_affiliation_string":"Tsinghua University,Shenzhen International Graduate School,Shenzhen,China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5076375643","display_name":"Yifei Zhao","orcid":"https://orcid.org/0000-0002-9614-4863"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yifei Zhao","raw_affiliation_strings":["Tsinghua University,Shenzhen International Graduate School,Shenzhen,China"],"affiliations":[{"raw_affiliation_string":"Tsinghua University,Shenzhen International Graduate School,Shenzhen,China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026500154","display_name":"Yongzhe Chang","orcid":null},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yongzhe Chang","raw_affiliation_strings":["Tsinghua University,Shenzhen International Graduate School,Shenzhen,China"],"affiliations":[{"raw_affiliation_string":"Tsinghua University,Shenzhen International Graduate School,Shenzhen,China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100737125","display_name":"Xueqian Wang","orcid":"https://orcid.org/0000-0003-3542-0593"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xueqian Wang","raw_affiliation_strings":["Tsinghua University,Shenzhen International Graduate School,Shenzhen,China"],"affiliations":[{"raw_affiliation_string":"Tsinghua University,Shenzhen International Graduate School,Shenzhen,China","institution_ids":["https://openalex.org/I99065089"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5100538111"],"corresponding_institution_ids":["https://openalex.org/I99065089"],"apc_list":null,"apc_paid":null,"fwci":4.1063,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.93120978,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":91,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"5"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10824","display_name":"Image Retrieval and Classification Techniques","score":0.9825000166893005,"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/T10824","display_name":"Image Retrieval and Classification Techniques","score":0.9825000166893005,"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/computer-science","display_name":"Computer science","score":0.6001554131507874},{"id":"https://openalex.org/keywords/preference","display_name":"Preference","score":0.5782602429389954},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.523855447769165},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4758607745170593},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.4323117136955261},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.1899470090866089},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.17382755875587463}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6001554131507874},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.5782602429389954},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.523855447769165},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4758607745170593},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.4323117136955261},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.1899470090866089},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.17382755875587463}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49660.2025.10890823","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10890823","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320321921","display_name":"Natural Science Foundation of Guangdong Province","ror":null},{"id":"https://openalex.org/F4320337504","display_name":"Research and Development","ror":"https://ror.org/027s68j25"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":20,"referenced_works":["https://openalex.org/W2138621090","https://openalex.org/W2917551568","https://openalex.org/W3035524453","https://openalex.org/W3153469116","https://openalex.org/W3175593095","https://openalex.org/W4312933868","https://openalex.org/W4390874002","https://openalex.org/W4402753770","https://openalex.org/W4402753790","https://openalex.org/W4409362098","https://openalex.org/W6774314701","https://openalex.org/W6778102432","https://openalex.org/W6779823529","https://openalex.org/W6846007759","https://openalex.org/W6852065897","https://openalex.org/W6852162230","https://openalex.org/W6852418670","https://openalex.org/W6852915159","https://openalex.org/W6853146131","https://openalex.org/W6861431061"],"related_works":["https://openalex.org/W2772917594","https://openalex.org/W2036807459","https://openalex.org/W2058170566","https://openalex.org/W2755342338","https://openalex.org/W2166024367","https://openalex.org/W3116076068","https://openalex.org/W2229312674","https://openalex.org/W2951359407","https://openalex.org/W2079911747","https://openalex.org/W1969923398"],"abstract_inverted_index":{"Direct":[0],"Preference":[1,100],"Optimization":[2],"(DPO)":[3],"has":[4],"recently":[5],"expanded":[6],"its":[7],"successful":[8],"application":[9],"beyond":[10],"aligning":[11],"large":[12],"language":[13],"models":[14,22],"(LLMs),":[15],"further":[16],"targeting":[17],"the":[18,46,70,81,97,119,142],"alignment":[19,47,156],"of":[20,37,63,72,117,122],"text-to-image":[21],"with":[23,146],"human":[24],"preferences.":[25],"However,":[26],"traditional":[27],"DPO":[28,64,78],"approach":[29],"would":[30],"inadvertently":[31],"result":[32],"in":[33,69,125],"a":[34,61],"simultaneous":[35],"reduction":[36],"sampling":[38],"probabilities":[39],"for":[40],"preferred":[41,123],"and":[42,109,134],"dispreferred":[43],"items":[44],"during":[45],"process,":[48],"thereby":[49],"potentially":[50],"diminishing":[51],"model's":[52],"generative":[53],"capacity.":[54],"In":[55],"this":[56],"paper,":[57],"we":[58,95,114],"firstly":[59],"undertake":[60],"revisit":[62],"by":[65,131],"grounding":[66],"our":[67],"analysis":[68],"framework":[71],"contrastive":[73],"loss.":[74],"It":[75],"reveals":[76],"that":[77,149],"only":[79],"emphasizes":[80],"part":[82],"quantifying":[83],"dissimilarity":[84],"between":[85],"items,":[86],"while":[87],"overlooking":[88],"aspects":[89],"pertinent":[90],"to":[91],"positive":[92],"items.":[93],"Hence,":[94],"propose":[96],"Positive":[98],"Enhanced":[99],"Alignment":[101],"(PEPA).":[102],"Three":[103],"enhancement":[104],"strategies":[105],"are":[106,138],"introduced":[107],"herein,":[108],"after":[110],"comprehensive":[111],"empirical":[112],"evaluation,":[113],"recommend":[115],"implementation":[116],"enhancing":[118],"log":[120],"probability":[121],"ratio":[124],"practice":[126],"applications,":[127],"which":[128],"is":[129],"distinguished":[130],"both":[132],"stability":[133],"effectiveness.":[135],"Experimental":[136],"assessments":[137],"carried":[139],"out":[140],"on":[141],"HPS-V2":[143],"test":[144],"set,":[145],"results":[147],"demonstrating":[148],"PEPA":[150],"outperforms":[151],"or":[152],"matches":[153],"current":[154],"state-of-the-art":[155],"techniques,":[157],"thus":[158],"highlighting":[159],"PEPA's":[160],"exceptional":[161],"practical":[162],"efficacy.":[163]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":1}],"updated_date":"2025-12-28T23:10:05.387466","created_date":"2025-10-10T00:00:00"}