{"id":"https://openalex.org/W4408352634","doi":"https://doi.org/10.1109/icassp49660.2025.10889679","title":"Minimizing Disparities between Real and Pseudo Queries for Unsupervised Visual Grounding","display_name":"Minimizing Disparities between Real and Pseudo Queries for Unsupervised Visual Grounding","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408352634","doi":"https://doi.org/10.1109/icassp49660.2025.10889679"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10889679","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10889679","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":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://figshare.com/articles/conference_contribution/Minimizing_Disparities_between_Real_and_Pseudo_Queries_for_Unsupervised_Visual_Grounding/29123075","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100742764","display_name":"Hui Jiang","orcid":"https://orcid.org/0000-0002-4534-1322"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hui Jiang","raw_affiliation_strings":["Fudan University,School of Computer Science, Shanghai Key Laboratory of Intelligent Information Processing,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University,School of Computer Science, Shanghai Key Laboratory of Intelligent Information Processing,Shanghai,China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111097611","display_name":"Changkai Ji","orcid":"https://orcid.org/0009-0007-7090-7360"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Changkai Ji","raw_affiliation_strings":["Fudan University,School of Computer Science, Shanghai Key Laboratory of Intelligent Information Processing,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University,School of Computer Science, Shanghai Key Laboratory of Intelligent Information Processing,Shanghai,China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5019996173","display_name":"Jilan Xu","orcid":"https://orcid.org/0000-0002-8208-2067"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jilan Xu","raw_affiliation_strings":["Fudan University,School of Computer Science, Shanghai Key Laboratory of Intelligent Information Processing,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University,School of Computer Science, Shanghai Key Laboratory of Intelligent Information Processing,Shanghai,China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5007003792","display_name":"Yanhao Zhu","orcid":null},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yanhao Zhu","raw_affiliation_strings":["Fudan University,School of Computer Science, Shanghai Key Laboratory of Intelligent Information Processing,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University,School of Computer Science, Shanghai Key Laboratory of Intelligent Information Processing,Shanghai,China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114236957","display_name":"Yuejie Zhang","orcid":"https://orcid.org/0009-0006-3746-6383"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuejie Zhang","raw_affiliation_strings":["Fudan University,School of Computer Science, Shanghai Key Laboratory of Intelligent Information Processing,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University,School of Computer Science, Shanghai Key Laboratory of Intelligent Information Processing,Shanghai,China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100680617","display_name":"Rui Feng","orcid":"https://orcid.org/0000-0002-4747-0574"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Rui Feng","raw_affiliation_strings":["Fudan University,School of Computer Science, Shanghai Key Laboratory of Intelligent Information Processing,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University,School of Computer Science, Shanghai Key Laboratory of Intelligent Information Processing,Shanghai,China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025376095","display_name":"Tao Zhang","orcid":"https://orcid.org/0000-0002-2801-8216"},"institutions":[{"id":"https://openalex.org/I181679659","display_name":"Shanghai University of Finance and Economics","ror":"https://ror.org/00wtvfq62","country_code":"CN","type":"education","lineage":["https://openalex.org/I181679659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tao Zhang","raw_affiliation_strings":["Shanghai University of Finance and Economics,School of Information Management and Engineering, Shanghai Key Laboratory of Financial Information Technology,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai University of Finance and Economics,School of Information Management and Engineering, Shanghai Key Laboratory of Financial Information Technology,Shanghai,China","institution_ids":["https://openalex.org/I181679659"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5102769979","display_name":"Shang Gao","orcid":"https://orcid.org/0000-0003-0135-6116"},"institutions":[{"id":"https://openalex.org/I149704539","display_name":"Deakin University","ror":"https://ror.org/02czsnj07","country_code":"AU","type":"education","lineage":["https://openalex.org/I149704539"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Shang Gao","raw_affiliation_strings":["Deakin University,School of Information Technology,Victoria,Australia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Deakin University,School of Information Technology,Victoria,Australia","institution_ids":["https://openalex.org/I149704539"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9891999959945679,"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":0.9891999959945679,"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":0.9865999817848206,"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/T10812","display_name":"Human Pose and Action Recognition","score":0.9639999866485596,"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":0.7447816133499146},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3876982629299164}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7447816133499146},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3876982629299164}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/icassp49660.2025.10889679","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10889679","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"},{"id":"pmh:oai:figshare.com:article/29123075","is_oa":true,"landing_page_url":"https://figshare.com/articles/conference_contribution/Minimizing_Disparities_between_Real_and_Pseudo_Queries_for_Unsupervised_Visual_Grounding/29123075","pdf_url":null,"source":{"id":"https://openalex.org/S4377196282","display_name":"Figshare","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I4210132348","host_organization_name":"Figshare (United Kingdom)","host_organization_lineage":["https://openalex.org/I4210132348"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Conference contribution"}],"best_oa_location":{"id":"pmh:oai:figshare.com:article/29123075","is_oa":true,"landing_page_url":"https://figshare.com/articles/conference_contribution/Minimizing_Disparities_between_Real_and_Pseudo_Queries_for_Unsupervised_Visual_Grounding/29123075","pdf_url":null,"source":{"id":"https://openalex.org/S4377196282","display_name":"Figshare","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I4210132348","host_organization_name":"Figshare (United Kingdom)","host_organization_lineage":["https://openalex.org/I4210132348"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Conference contribution"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320321885","display_name":"Science and Technology Commission of Shanghai Municipality","ror":"https://ror.org/03kt66j61"},{"id":"https://openalex.org/F4320335787","display_name":"Fundamental Research Funds for the Central Universities","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":40,"referenced_works":["https://openalex.org/W1773149199","https://openalex.org/W1861492603","https://openalex.org/W2251512949","https://openalex.org/W2489434015","https://openalex.org/W2570343428","https://openalex.org/W2745461083","https://openalex.org/W2948672349","https://openalex.org/W2962764817","https://openalex.org/W2963109634","https://openalex.org/W2963614783","https://openalex.org/W2964345792","https://openalex.org/W2981663434","https://openalex.org/W2984121207","https://openalex.org/W2984194315","https://openalex.org/W2987401211","https://openalex.org/W3034655362","https://openalex.org/W3110435696","https://openalex.org/W3117585461","https://openalex.org/W3126391825","https://openalex.org/W3163747765","https://openalex.org/W3178418424","https://openalex.org/W3179041377","https://openalex.org/W4214490042","https://openalex.org/W4225517085","https://openalex.org/W4226104391","https://openalex.org/W4235505822","https://openalex.org/W4285192809","https://openalex.org/W4313145013","https://openalex.org/W4366352717","https://openalex.org/W4387272106","https://openalex.org/W4390873208","https://openalex.org/W4391451889","https://openalex.org/W4400721967","https://openalex.org/W4404612934","https://openalex.org/W6620707391","https://openalex.org/W6678224444","https://openalex.org/W6797397777","https://openalex.org/W6803567076","https://openalex.org/W6851592950","https://openalex.org/W6851950068"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"Visual":[0],"grounding":[1,24,36,63,75],"involves":[2],"the":[3,15,58,91,95,116],"identification":[4],"and":[5,31,45,54,101,123],"localization":[6],"of":[7,50,61,118],"image":[8],"regions":[9],"given":[10],"textual":[11,55,119],"descriptions.":[12],"To":[13,65],"reduce":[14],"manual":[16],"labeling":[17],"effort":[18],"on":[19,137],"region-text":[20],"pairs,":[21],"unsupervised":[22,62,73,135],"visual":[23,74],"aims":[25],"to":[26,85,114],"generate":[27,86],"pseudo":[28,46,87,124],"bounding":[29],"box":[30],"query":[32],"pairs":[33],"for":[34],"training":[35],"models.":[37],"However,":[38],"there":[39],"exists":[40],"significant":[41],"disparities":[42],"between":[43,121],"real":[44,122],"queries":[47],"in":[48,89],"terms":[49],"object,":[51],"attribute":[52],"distributions,":[53],"formats,":[56],"limiting":[57],"generalization":[59],"performance":[60],"methods.":[64],"address":[66],"this":[67],"challenge,":[68],"we":[69,79],"propose":[70],"a":[71,110],"novel":[72],"framework.":[76],"During":[77],"training,":[78],"prompt":[80],"Multimodal":[81],"Large":[82],"Language":[83],"Models":[84],"queries,":[88],"which":[90],"entities":[92],"are":[93,102],"beyond":[94],"object":[96],"detector\u2019s":[97],"pre-defined":[98],"limited":[99],"categories,":[100],"associated":[103],"with":[104,144],"richer":[105],"attributes.":[106],"We":[107],"further":[108],"devise":[109],"Modifier":[111],"Tree":[112],"structure":[113],"bridge":[115],"gap":[117],"format":[120],"queries.":[125,146],"Extensive":[126],"experiments":[127],"demonstrate":[128],"that":[129],"our":[130],"method":[131],"significantly":[132],"outperforms":[133],"state-of-the-art":[134],"approaches":[136],"public":[138],"benchmark":[139],"datasets,":[140],"particularly":[141],"when":[142],"dealing":[143],"complex":[145]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}