{"id":"https://openalex.org/W7160332232","doi":"https://doi.org/10.1109/wacv61042.2026.00829","title":"Relevance-aware Multi-context Contrastive Decoding for Retrieval-augmented Visual Question Answering","display_name":"Relevance-aware Multi-context Contrastive Decoding for Retrieval-augmented Visual Question Answering","publication_year":2026,"publication_date":"2026-03-06","ids":{"openalex":"https://openalex.org/W7160332232","doi":"https://doi.org/10.1109/wacv61042.2026.00829"},"language":null,"primary_location":{"id":"doi:10.1109/wacv61042.2026.00829","is_oa":false,"landing_page_url":"https://doi.org/10.1109/wacv61042.2026.00829","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5135361896","display_name":"Jongha Kim","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jongha Kim","raw_affiliation_strings":["Korea University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea University","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110502130","display_name":"B. Ko","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Byungoh Ko","raw_affiliation_strings":["Korea University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea University","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5084675431","display_name":"Jeehye Na","orcid":null},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]},{"id":"https://openalex.org/I4210099236","display_name":"Kootenay Association for Science & Technology","ror":"https://ror.org/011pv9p44","country_code":"CA","type":"nonprofit","lineage":["https://openalex.org/I4210099236"]}],"countries":["CA","KR"],"is_corresponding":false,"raw_author_name":"Jeehye Na","raw_affiliation_strings":["KAIST"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST","institution_ids":["https://openalex.org/I157485424","https://openalex.org/I4210099236"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135341002","display_name":"Jinsung Yoon","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jinsung Yoon","raw_affiliation_strings":["Google Cloud AI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google Cloud AI","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5135300637","display_name":"Hyunwoo J. Kim","orcid":null},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]},{"id":"https://openalex.org/I4210099236","display_name":"Kootenay Association for Science & Technology","ror":"https://ror.org/011pv9p44","country_code":"CA","type":"nonprofit","lineage":["https://openalex.org/I4210099236"]}],"countries":["CA","KR"],"is_corresponding":false,"raw_author_name":"Hyunwoo J. Kim","raw_affiliation_strings":["KAIST"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST","institution_ids":["https://openalex.org/I157485424","https://openalex.org/I4210099236"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"8596","last_page":"8605"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9312000274658203,"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":0.9312000274658203,"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":0.021199999377131462,"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"}},{"id":"https://openalex.org/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.007699999958276749,"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/question-answering","display_name":"Question answering","score":0.7077000141143799},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.4505000114440918},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.3098999857902527},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.30709999799728394},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.2856999933719635}],"concepts":[{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.7077000141143799},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6330999732017517},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5652999877929688},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5568000078201294},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.4505000114440918},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.32499998807907104},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.3098999857902527},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.30709999799728394},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.2856999933719635},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.2702000141143799},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.267300009727478}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/wacv61042.2026.00829","is_oa":false,"landing_page_url":"https://doi.org/10.1109/wacv61042.2026.00829","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320322120","display_name":"National Research Foundation of Korea","ror":"https://ror.org/013aysd81"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":29,"referenced_works":["https://openalex.org/W1861492603","https://openalex.org/W2560730294","https://openalex.org/W2904565150","https://openalex.org/W2947312908","https://openalex.org/W2963518342","https://openalex.org/W2966317026","https://openalex.org/W3135279933","https://openalex.org/W3196798856","https://openalex.org/W4226321975","https://openalex.org/W4385571645","https://openalex.org/W4385574177","https://openalex.org/W4386065596","https://openalex.org/W4386065803","https://openalex.org/W4389523839","https://openalex.org/W4389524050","https://openalex.org/W4390871810","https://openalex.org/W4390872646","https://openalex.org/W4402672071","https://openalex.org/W4402704565","https://openalex.org/W4402727764","https://openalex.org/W4402753774","https://openalex.org/W4402753980","https://openalex.org/W4402916777","https://openalex.org/W4404782107","https://openalex.org/W4409368434","https://openalex.org/W7133193597","https://openalex.org/W7133220561","https://openalex.org/W7138058965","https://openalex.org/W7160018935"],"related_works":[],"abstract_inverted_index":{"Despite":[0],"the":[1,60,104,121,140,157,173,178,181,184],"remarkable":[2],"capabilities":[3],"of":[4,63,124,160],"Large":[5],"Vision":[6],"Language":[7],"Models":[8],"(LVLMs),":[9],"they":[10,50],"still":[11],"lack":[12],"detailed":[13],"knowledge":[14],"about":[15],"specific":[16],"entities.":[17],"Retrieval-augmented":[18],"Generation":[19],"(RAG)":[20],"is":[21,97,170],"a":[22,76,84],"widely":[23],"adopted":[24],"solution":[25],"that":[26,41,129,168],"enhances":[27],"LVLMs":[28,161],"by":[29,87,155],"providing":[30],"additional":[31,163],"contexts":[32,57,117],"from":[33,114],"an":[34],"external":[35],"Knowledge":[36],"Base.":[37],"However,":[38],"we":[39,69],"observe":[40],"previous":[42],"decoding":[43,78,134,158],"methods":[44,135],"for":[45,80],"RAG":[46],"are":[47],"sub-optimal":[48],"as":[49],"fail":[51],"to":[52,103,172,183],"sufficiently":[53],"leverage":[54],"multiple":[55,115,137],"relevant":[56,116],"and":[58],"suppress":[59],"negative":[61,122],"effects":[62,123],"irrelevant":[64,125],"contexts.":[65],"To":[66],"this":[67],"end,":[68],"propose":[70],"Relevance-aware":[71],"Multi-context":[72],"Contrastive":[73],"Decoding":[74],"(RMCD),":[75],"novel":[77],"method":[79,159],"RAG.":[81],"RMCD":[82,109,130,150,169],"outputs":[83,89],"final":[85],"prediction":[86],"combining":[88],"predicted":[90],"with":[91],"each":[92,95],"context,":[93],"where":[94],"output":[96],"weighted":[98],"based":[99],"on":[100,143],"its":[101],"relevance":[102],"question.":[105],"By":[106],"doing":[107],"so,":[108],"effectively":[110],"aggregates":[111],"useful":[112],"information":[113],"while":[118],"also":[119,166],"counteracting":[120],"ones.":[126],"Experiments":[127],"show":[128,167],"consistently":[131,176],"outperforms":[132],"other":[133],"across":[136,180],"LVLMs,":[138],"achieving":[139],"best":[141,179],"performance":[142],"three":[144],"knowledge-intensive":[145],"visual":[146],"question-answering":[147],"benchmarks.":[148],"Also,":[149],"can":[151],"be":[152],"simply":[153],"applied":[154],"replacing":[156],"without":[162],"training.":[164],"Analyses":[165],"robust":[171],"retrieval":[174,186],"results,":[175],"performing":[177],"weakest":[182],"strongest":[185],"results.":[187]},"counts_by_year":[],"updated_date":"2026-07-22T07:51:19.307946","created_date":"2026-05-06T00:00:00"}