{"id":"https://openalex.org/W4392909493","doi":"https://doi.org/10.1109/icassp48485.2024.10446378","title":"Enhancing Generalization in Medical Visual Question Answering Tasks Via Gradient-Guided Model Perturbation","display_name":"Enhancing Generalization in Medical Visual Question Answering Tasks Via Gradient-Guided Model Perturbation","publication_year":2024,"publication_date":"2024-03-18","ids":{"openalex":"https://openalex.org/W4392909493","doi":"https://doi.org/10.1109/icassp48485.2024.10446378"},"language":"en","primary_location":{"id":"doi:10.1109/icassp48485.2024.10446378","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/icassp48485.2024.10446378","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 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/A5109282686","display_name":"Gang Liu","orcid":"https://orcid.org/0000-0002-7032-8429"},"institutions":[{"id":"https://openalex.org/I151727225","display_name":"Harbin Engineering University","ror":"https://ror.org/03x80pn82","country_code":"CN","type":"education","lineage":["https://openalex.org/I151727225"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Gang Liu","raw_affiliation_strings":["Harbin Engineering University,College of Computer Science and Technology,China","College of Computer Science and Technology, Harbin Engineering University, China"],"affiliations":[{"raw_affiliation_string":"Harbin Engineering University,College of Computer Science and Technology,China","institution_ids":["https://openalex.org/I151727225"]},{"raw_affiliation_string":"College of Computer Science and Technology, Harbin Engineering University, China","institution_ids":["https://openalex.org/I151727225"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5115594229","display_name":"Hongyang Li","orcid":null},"institutions":[{"id":"https://openalex.org/I151727225","display_name":"Harbin Engineering University","ror":"https://ror.org/03x80pn82","country_code":"CN","type":"education","lineage":["https://openalex.org/I151727225"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hongyang Li","raw_affiliation_strings":["Harbin Engineering University,College of Computer Science and Technology,China","College of Computer Science and Technology, Harbin Engineering University, China"],"affiliations":[{"raw_affiliation_string":"Harbin Engineering University,College of Computer Science and Technology,China","institution_ids":["https://openalex.org/I151727225"]},{"raw_affiliation_string":"College of Computer Science and Technology, Harbin Engineering University, China","institution_ids":["https://openalex.org/I151727225"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101319287","display_name":"Zerui He","orcid":null},"institutions":[{"id":"https://openalex.org/I151727225","display_name":"Harbin Engineering University","ror":"https://ror.org/03x80pn82","country_code":"CN","type":"education","lineage":["https://openalex.org/I151727225"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zerui He","raw_affiliation_strings":["Harbin Engineering University,College of Computer Science and Technology,China","College of Computer Science and Technology, Harbin Engineering University, China"],"affiliations":[{"raw_affiliation_string":"Harbin Engineering University,College of Computer Science and Technology,China","institution_ids":["https://openalex.org/I151727225"]},{"raw_affiliation_string":"College of Computer Science and Technology, Harbin Engineering University, China","institution_ids":["https://openalex.org/I151727225"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5006400852","display_name":"Shenjun Zhong","orcid":null},"institutions":[{"id":"https://openalex.org/I56590836","display_name":"Monash University","ror":"https://ror.org/02bfwt286","country_code":"AU","type":"education","lineage":["https://openalex.org/I56590836"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Shenjun Zhong","raw_affiliation_strings":["Monash University,Monash Biomedical Imaging,Australia","Monash Biomedical Imaging, Monash University, Australia"],"affiliations":[{"raw_affiliation_string":"Monash University,Monash Biomedical Imaging,Australia","institution_ids":["https://openalex.org/I56590836"]},{"raw_affiliation_string":"Monash Biomedical Imaging, Monash University, Australia","institution_ids":["https://openalex.org/I56590836"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5109282686"],"corresponding_institution_ids":["https://openalex.org/I151727225"],"apc_list":null,"apc_paid":null,"fwci":0.2501,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.46391046,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"2220","last_page":"2224"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9764000177383423,"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":0.9764000177383423,"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/T10824","display_name":"Image Retrieval and Classification Techniques","score":0.95169997215271,"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":0.6873955726623535},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.65854811668396},{"id":"https://openalex.org/keywords/question-answering","display_name":"Question answering","score":0.5801241993904114},{"id":"https://openalex.org/keywords/perturbation","display_name":"Perturbation (astronomy)","score":0.5338550806045532},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3895007371902466},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.13035738468170166}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6873955726623535},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.65854811668396},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.5801241993904114},{"id":"https://openalex.org/C177918212","wikidata":"https://www.wikidata.org/wiki/Q803623","display_name":"Perturbation (astronomy)","level":2,"score":0.5338550806045532},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3895007371902466},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.13035738468170166},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp48485.2024.10446378","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/icassp48485.2024.10446378","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/13","display_name":"Climate action","score":0.6200000047683716}],"awards":[{"id":"https://openalex.org/G4023856508","display_name":null,"funder_award_id":"LH2021F015","funder_id":"https://openalex.org/F4320323085","funder_display_name":"Natural Science Foundation of Heilongjiang Province"}],"funders":[{"id":"https://openalex.org/F4320321408","display_name":"Ministry of Education","ror":"https://ror.org/01p262204"},{"id":"https://openalex.org/F4320323085","display_name":"Natural Science Foundation of Heilongjiang Province","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":22,"referenced_works":["https://openalex.org/W1522301498","https://openalex.org/W2095705004","https://openalex.org/W2897980926","https://openalex.org/W2901466771","https://openalex.org/W2963485691","https://openalex.org/W2979525559","https://openalex.org/W3035524453","https://openalex.org/W3165058054","https://openalex.org/W3203635737","https://openalex.org/W4256202811","https://openalex.org/W4283258232","https://openalex.org/W4293676823","https://openalex.org/W4295917908","https://openalex.org/W4313452817","https://openalex.org/W4377121462","https://openalex.org/W4386352883","https://openalex.org/W6631190155","https://openalex.org/W6674330103","https://openalex.org/W6747657013","https://openalex.org/W6784333009","https://openalex.org/W6798805250","https://openalex.org/W6852751528"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2748952813","https://openalex.org/W2384605597","https://openalex.org/W2387743295","https://openalex.org/W2115758952","https://openalex.org/W3082787378","https://openalex.org/W2136007095","https://openalex.org/W2366230879","https://openalex.org/W3208425359","https://openalex.org/W2349927912"],"abstract_inverted_index":{"Leveraging":[0],"pre-trained":[1],"visual":[2,16,100,159],"language":[3],"models":[4],"has":[5],"become":[6],"a":[7,35,91,167],"widely":[8],"adopted":[9],"approach":[10,175],"for":[11,116],"improving":[12],"performance":[13],"in":[14,22,137],"downstream":[15,117],"question":[17],"answering":[18],"(VQA)":[19],"applications.":[20],"However,":[21],"the":[23,29,53,68,99,103,130,133,138,145,148,157],"specialized":[24],"field":[25],"of":[26,31,102,132,147],"medical":[27,118],"VQA,":[28],"scarcity":[30],"available":[32],"data":[33,84],"poses":[34],"significant":[36],"barrier":[37],"to":[38,48,66,75,98,112,144],"achieving":[39],"reliable":[40],"model":[41,50,77,105,114],"generalization.":[42],"Numerous":[43],"methods":[44],"have":[45],"been":[46],"proposed":[47],"enhance":[49],"generalization,":[51],"addressing":[52],"issue":[54],"from":[55],"data-centric":[56],"and":[57,109,182],"model-centric":[58],"perspectives.":[59],"Data":[60],"augmentation":[61],"techniques":[62],"are":[63],"commonly":[64],"employed":[65],"enrich":[67],"dataset,":[69,173],"while":[70],"various":[71],"regularization":[72],"approaches":[73],"aim":[74],"prevent":[76],"overfitting,":[78],"especially":[79],"when":[80],"training":[81],"on":[82,179],"limited":[83],"samples.":[85],"In":[86],"this":[87],"paper,":[88],"we":[89],"introduce":[90],"method":[92],"that":[93],"incorporates":[94],"gradient-guided":[95],"parameter":[96],"perturbations":[97],"encoder":[101],"multimodality":[104],"during":[106],"both":[107,180],"pre-training":[108,170],"fine-tuning":[110],"phases,":[111],"improve":[113],"generalization":[115],"VQA":[119],"tasks.":[120],"The":[121,161],"small":[122],"perturbation":[123],"is":[124,142,153],"adaptively":[125],"generated":[126],"by":[127],"aligning":[128],"with":[129,166],"direction":[131],"moving":[134],"average":[135],"gradient":[136],"optimization":[139],"landscape,":[140],"which":[141],"opposite":[143],"directions":[146],"optimizer\u2019s":[149],"historical":[150],"updates.":[151],"It":[152],"subsequently":[154],"injected":[155],"into":[156],"model\u2019s":[158],"encoder.":[160],"results":[162],"show":[163],"that,":[164],"even":[165],"significantly":[168],"smaller":[169],"image":[171],"caption":[172],"our":[174],"achieves":[176],"competitive":[177],"outcomes":[178],"VQA-RAD":[181],"SLAKE":[183],"datasets.":[184]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-03-25T13:04:00.132906","created_date":"2025-10-10T00:00:00"}