{"id":"https://openalex.org/W4402753327","doi":"https://doi.org/10.1109/cvpr52733.2024.02660","title":"Groupwise Query Specialization and Quality-Aware Multi-Assignment for Transformer-Based Visual Relationship Detection","display_name":"Groupwise Query Specialization and Quality-Aware Multi-Assignment for Transformer-Based Visual Relationship Detection","publication_year":2024,"publication_date":"2024-06-16","ids":{"openalex":"https://openalex.org/W4402753327","doi":"https://doi.org/10.1109/cvpr52733.2024.02660"},"language":"en","primary_location":{"id":"doi:10.1109/cvpr52733.2024.02660","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cvpr52733.2024.02660","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","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/A5001493781","display_name":"Jongha Kim","orcid":null},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jongha Kim","raw_affiliation_strings":["Korea University,Department of Computer Science and Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea University,Department of Computer Science and Engineering","institution_ids":["https://openalex.org/I197347611"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100740582","display_name":"Jihwan Park","orcid":"https://orcid.org/0000-0002-9882-0482"},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jihwan Park","raw_affiliation_strings":["Korea University,Department of Computer Science and Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea University,Department of Computer Science and Engineering","institution_ids":["https://openalex.org/I197347611"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100428898","display_name":"Jinyoung Park","orcid":"https://orcid.org/0000-0003-3962-5668"},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jinyoung Park","raw_affiliation_strings":["Korea University,Department of Computer Science and Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea University,Department of Computer Science and Engineering","institution_ids":["https://openalex.org/I197347611"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100417174","display_name":"Jin Young Kim","orcid":"https://orcid.org/0000-0002-4896-8980"},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jinyoung Kim","raw_affiliation_strings":["Korea University,Department of Computer Science and Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea University,Department of Computer Science and Engineering","institution_ids":["https://openalex.org/I197347611"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043210468","display_name":"Sehyung Kim","orcid":"https://orcid.org/0009-0003-6381-9291"},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Sehyung Kim","raw_affiliation_strings":["Korea University,Department of Computer Science and Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea University,Department of Computer Science and Engineering","institution_ids":["https://openalex.org/I197347611"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5084814930","display_name":"Hyunwoo J. Kim","orcid":"https://orcid.org/0000-0002-2181-9264"},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Hyunwoo J. Kim","raw_affiliation_strings":["Korea University,Department of Computer Science and Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea University,Department of Computer Science and Engineering","institution_ids":["https://openalex.org/I197347611"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I197347611"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":11,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"28160","last_page":"28169"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9997000098228455,"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":0.9997000098228455,"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":0.9988999962806702,"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":0.9972000122070312,"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/computer-science","display_name":"Computer science","score":0.8190085291862488},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.7059965133666992},{"id":"https://openalex.org/keywords/quality","display_name":"Quality (philosophy)","score":0.4717859923839569},{"id":"https://openalex.org/keywords/query-optimization","display_name":"Query optimization","score":0.46381279826164246},{"id":"https://openalex.org/keywords/data-mining","display_name":"Data mining","score":0.3851964473724365},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"Information retrieval","score":0.32277965545654297},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3209162950515747},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.07345002889633179},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.0669504702091217},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.0619661808013916}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8190085291862488},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.7059965133666992},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.4717859923839569},{"id":"https://openalex.org/C157692150","wikidata":"https://www.wikidata.org/wiki/Q2919848","display_name":"Query optimization","level":2,"score":0.46381279826164246},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3851964473724365},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.32277965545654297},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3209162950515747},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.07345002889633179},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0669504702091217},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0619661808013916},{"id":"https://openalex.org/C111472728","wikidata":"https://www.wikidata.org/wiki/Q9471","display_name":"Epistemology","level":1,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/cvpr52733.2024.02660","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cvpr52733.2024.02660","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":47,"referenced_works":["https://openalex.org/W2077069816","https://openalex.org/W2194775991","https://openalex.org/W2222512263","https://openalex.org/W2277195237","https://openalex.org/W2579549467","https://openalex.org/W2962779575","https://openalex.org/W2963514444","https://openalex.org/W2963536419","https://openalex.org/W2963938081","https://openalex.org/W2964225075","https://openalex.org/W2969679616","https://openalex.org/W3035017890","https://openalex.org/W3081642947","https://openalex.org/W3096609285","https://openalex.org/W3106121867","https://openalex.org/W3106328333","https://openalex.org/W3108864070","https://openalex.org/W3168488421","https://openalex.org/W3171162369","https://openalex.org/W3171169846","https://openalex.org/W3174164794","https://openalex.org/W3174480456","https://openalex.org/W3174865181","https://openalex.org/W3195421894","https://openalex.org/W3196936439","https://openalex.org/W4214693531","https://openalex.org/W4225868495","https://openalex.org/W4287728378","https://openalex.org/W4304080274","https://openalex.org/W4312446811","https://openalex.org/W4312513541","https://openalex.org/W4312563197","https://openalex.org/W4312682661","https://openalex.org/W4312848112","https://openalex.org/W4312960102","https://openalex.org/W4313037583","https://openalex.org/W4313155487","https://openalex.org/W4386065740","https://openalex.org/W4390871882","https://openalex.org/W6739901393","https://openalex.org/W6757817989","https://openalex.org/W6775827216","https://openalex.org/W6780471797","https://openalex.org/W6785527955","https://openalex.org/W6793175614","https://openalex.org/W6799822930","https://openalex.org/W6840398736"],"related_works":["https://openalex.org/W2006459955","https://openalex.org/W1819934925","https://openalex.org/W4386051213","https://openalex.org/W2385922226","https://openalex.org/W2955368753","https://openalex.org/W2413368787","https://openalex.org/W2393004407","https://openalex.org/W2362665752","https://openalex.org/W2337762748","https://openalex.org/W185198413"],"abstract_inverted_index":{"Visual":[0],"Relationship":[1],"Detection":[2],"(VRD)":[3],"has":[4],"seen":[5],"significant":[6],"advancements":[7],"with":[8,205],"Transformer-based":[9,25],"architectures":[10],"recently.":[11],"However,":[12],"we":[13,108],"identify":[14],"two":[15],"key":[16],"limitations":[17],"in":[18,67,136,144,169,179,201],"a":[19,30,34,38,50,63,71,78,84,102,121,134,137,157,167,172,198],"conventional":[20,42],"label":[21],"assignment":[22],"for":[23,62],"training":[24,154],"VRD":[26,212],"models,":[27],"which":[28,58,192],"is":[29,47,52,73,80,217],"process":[31],"of":[32,171,197],"mapping":[33],"ground-truth":[35],"(GT)":[36],"to":[37,54,65,83,159,166],"prediction.":[39],"Under":[40],"the":[41,145,153,177,195],"assignment,":[43],"an":[44,174],"\u2018unspecialized\u2019":[45],"query":[46,51,64,72,123,135,139],"trained":[48,76],"since":[49,77],"expected":[53],"detect":[55],"every":[56],"relation,":[57],"makes":[59],"it":[60],"difficult":[61],"specialize":[66],"specific":[68,138],"relations.":[69],"Furthermore,":[70],"also":[74],"insufficiently":[75],"GT":[79,158,168],"assigned":[81,97],"only":[82],"single":[85],"prediction,":[86],"therefore":[87],"near-correct":[88],"or":[89],"even":[90],"correct":[91],"predictions":[92,161],"are":[93,163],"suppressed":[94],"by":[95,124,155],"being":[96],"\u2018no":[98],"relation":[99,147,178],"(\u2298)\u2018":[100],"as":[101],"GT.":[103],"To":[104],"address":[105],"these":[106],"issues,":[107],"propose":[109],"Groupwise":[110,117],"Query":[111,118],"Specialization":[112,119],"and":[113,127,132,176,183,214],"Quality-Aware":[114,149],"Multi-Assignment":[115,150],"(SpeaQ).":[116],"trains":[120,189],"\u2018specialized\u2019":[122,190],"dividing":[125],"queries":[126],"relations":[128,143],"into":[129],"disjoint":[130],"groups":[131],"directing":[133],"group":[140],"solely":[141],"toward":[142],"corresponding":[146],"group.":[148],"further":[151],"facilitates":[152],"assigning":[156],"multiple":[160,211],"that":[162,186],"significantly":[164],"close":[165],"terms":[170],"subject,":[173],"object,":[175],"between.":[180],"Experimental":[181],"results":[182],"analyses":[184],"show":[185],"SpeaQ":[187],"effectively":[188],"queries,":[191],"better":[193],"utilize":[194],"capacity":[196],"model,":[199],"resulting":[200],"consistent":[202],"performance":[203],"gains":[204],"\u2018zero\u2019":[206],"additional":[207],"inference":[208],"cost":[209],"across":[210],"models":[213],"benchmarks.":[215],"Code":[216],"available":[218],"at":[219],"https://github.com/m1vlab/SpeaQ.":[220]},"counts_by_year":[{"year":2026,"cited_by_count":4},{"year":2025,"cited_by_count":7}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}