{"id":"https://openalex.org/W4403791078","doi":"https://doi.org/10.1145/3664647.3680993","title":"Self-Supervised Visual Preference Alignment","display_name":"Self-Supervised Visual Preference Alignment","publication_year":2024,"publication_date":"2024-10-26","ids":{"openalex":"https://openalex.org/W4403791078","doi":"https://doi.org/10.1145/3664647.3680993"},"language":"en","primary_location":{"id":"doi:10.1145/3664647.3680993","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3664647.3680993","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 32nd ACM International Conference on Multimedia","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/A5100962035","display_name":"Ke Zhu","orcid":null},"institutions":[{"id":"https://openalex.org/I881766915","display_name":"Nanjing University","ror":"https://ror.org/01rxvg760","country_code":"CN","type":"education","lineage":["https://openalex.org/I881766915"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Ke Zhu","raw_affiliation_strings":["School of Artificial Intelligence, Nanjing University, Nanjing, China"],"raw_orcid":"https://orcid.org/0009-0007-6338-4151","affiliations":[{"raw_affiliation_string":"School of Artificial Intelligence, Nanjing University, Nanjing, China","institution_ids":["https://openalex.org/I881766915"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5115603829","display_name":"Liang Zhao","orcid":"https://orcid.org/0000-0002-4191-7814"},"institutions":[{"id":"https://openalex.org/I4401726805","display_name":"Megvii (China)","ror":"https://ror.org/040b32p69","country_code":null,"type":"company","lineage":["https://openalex.org/I4401726805"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Liang Zhao","raw_affiliation_strings":["Megvii Technology, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-4191-7814","affiliations":[{"raw_affiliation_string":"Megvii Technology, Beijing, China","institution_ids":["https://openalex.org/I4401726805"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5106405555","display_name":"Zheng Ge","orcid":"https://orcid.org/0000-0002-8630-8270"},"institutions":[{"id":"https://openalex.org/I4401726805","display_name":"Megvii (China)","ror":"https://ror.org/040b32p69","country_code":null,"type":"company","lineage":["https://openalex.org/I4401726805"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zheng Ge","raw_affiliation_strings":["Megvii Technology, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-8630-8270","affiliations":[{"raw_affiliation_string":"Megvii Technology, Beijing, China","institution_ids":["https://openalex.org/I4401726805"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100362465","display_name":"Xiangyu Zhang","orcid":"https://orcid.org/0000-0003-2138-4608"},"institutions":[{"id":"https://openalex.org/I4401726805","display_name":"Megvii (China)","ror":"https://ror.org/040b32p69","country_code":null,"type":"company","lineage":["https://openalex.org/I4401726805"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiangyu Zhang","raw_affiliation_strings":["Megvii Technology, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-2138-4608","affiliations":[{"raw_affiliation_string":"Megvii Technology, Beijing, China","institution_ids":["https://openalex.org/I4401726805"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5100962035"],"corresponding_institution_ids":["https://openalex.org/I881766915"],"apc_list":null,"apc_paid":null,"fwci":0.9523,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":{"value":0.77049561,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":97,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"291","last_page":"300"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","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/T11714","display_name":"Multimodal Machine Learning Applications","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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9994999766349792,"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":0.9973000288009644,"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/preference","display_name":"Preference","score":0.6934481859207153},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6599228382110596},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5085924863815308},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.38543814420700073},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.12762179970741272},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.09436887502670288}],"concepts":[{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.6934481859207153},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6599228382110596},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5085924863815308},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.38543814420700073},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.12762179970741272},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.09436887502670288}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3664647.3680993","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3664647.3680993","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 32nd ACM International Conference on Multimedia","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":12,"referenced_works":["https://openalex.org/W2277195237","https://openalex.org/W2963518342","https://openalex.org/W2979382951","https://openalex.org/W3004268082","https://openalex.org/W4212774754","https://openalex.org/W4376312115","https://openalex.org/W4379806236","https://openalex.org/W4379806273","https://openalex.org/W4379806303","https://openalex.org/W4389523832","https://openalex.org/W4390873788","https://openalex.org/W4393159845"],"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":{"This":[0],"paper":[1],"makes":[2],"the":[3,23,48,63,147,151],"first":[4],"attempt":[5],"towards":[6,158],"unsupervised":[7,104],"preference":[8,31,35],"alignment":[9,32],"in":[10,116],"Vision-Language":[11],"Models":[12],"(VLMs).":[13],"We":[14],"generate":[15,55],"chosen":[16],"and":[17,25,29,68,72,90,118],"rejected":[18],"responses":[19],"with":[20,33,94,136],"regard":[21],"to":[22,47,54,65,111,134,145],"original":[24],"augmented":[26],"image":[27,49],"pairs,":[28],"conduct":[30],"direct":[34],"optimization.":[36],"It":[37],"is":[38,91],"based":[39],"on":[40,81,113,124],"a":[41],"core":[42],"idea:":[43],"properly":[44],"designed":[45],"augmentation":[46],"input":[50],"will":[51],"induce":[52],"VLM":[53],"false":[56],"but":[57],"hard":[58],"negative":[59],"responses,":[60],"which":[61,153],"helps":[62],"model":[64],"learn":[66],"from":[67,83],"produce":[69],"more":[70],"robust":[71],"powerful":[73],"answers.":[74],"The":[75],"whole":[76],"pipeline":[77],"no":[78],"longer":[79],"hinges":[80],"supervision":[82],"GPT-4":[84,112],"or":[85],"human":[86],"involvement":[87],"during":[88],"alignment,":[89],"highly":[92],"efficient":[93],"few":[95],"lines":[96],"of":[97,140,150],"code.":[98],"With":[99],"only":[100],"8k":[101],"randomly":[102],"sampled":[103],"data,":[105],"it":[106],"achieves":[107],"90%":[108],"relative":[109],"score":[110,123],"complex":[114,125],"reasoning":[115],"LLaVA-Bench,":[117],"improves":[119],"LLaVA-7B/13B":[120],"by":[121],"6.7%/5.6%":[122],"multi-modal":[126],"benchmark":[127],"MM-Vet.":[128],"Visualizations":[129],"shows":[130],"its":[131,156],"improved":[132],"ability":[133],"align":[135],"user-intentions.":[137],"A":[138],"series":[139],"ablations":[141],"are":[142],"firmly":[143],"conducted":[144],"reveal":[146],"latent":[148],"mechanism":[149],"approach,":[152],"also":[154],"indicates":[155],"potential":[157],"further":[159],"scaling.":[160]},"counts_by_year":[{"year":2025,"cited_by_count":4}],"updated_date":"2025-12-21T01:58:51.020947","created_date":"2025-10-10T00:00:00"}