{"id":"https://openalex.org/W7161778661","doi":"https://doi.org/10.48550/arxiv.2605.19307","title":"MetaRA: Metamorphic Robustness Assessment for Multimodal Large Language Model-based Visual Question Answering Systems","display_name":"MetaRA: Metamorphic Robustness Assessment for Multimodal Large Language Model-based Visual Question Answering Systems","publication_year":2026,"publication_date":"2026-05-19","ids":{"openalex":"https://openalex.org/W7161778661","doi":"https://doi.org/10.48550/arxiv.2605.19307"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.19307","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.19307","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.19307","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5136571811","display_name":"Quanxing Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Quanxing","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5106406774","display_name":"Ye Tian","orcid":"https://orcid.org/0000-0002-5495-6492"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tian, Yuhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136580665","display_name":"Ling Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Ling","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136549124","display_name":"Xian Zhong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhong, Xian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136611095","display_name":"Xiaohua Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Xiaohua","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136548730","display_name":"Rubing Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Rubing","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136522040","display_name":"Chia-Wen Lin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lin, Chia-Wen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"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":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.8787000179290771,"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":0.8787000179290771,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.027899999171495438,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.022600000724196434,"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/robustness","display_name":"Robustness (evolution)","score":0.8001000285148621},{"id":"https://openalex.org/keywords/question-answering","display_name":"Question answering","score":0.6237000226974487},{"id":"https://openalex.org/keywords/trustworthiness","display_name":"Trustworthiness","score":0.5266000032424927},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.4562000036239624},{"id":"https://openalex.org/keywords/computational-model","display_name":"Computational model","score":0.3522999882698059},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.3098999857902527}],"concepts":[{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.8001000285148621},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.75},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.6237000226974487},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5738000273704529},{"id":"https://openalex.org/C153701036","wikidata":"https://www.wikidata.org/wiki/Q659974","display_name":"Trustworthiness","level":2,"score":0.5266000032424927},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.48069998621940613},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.4562000036239624},{"id":"https://openalex.org/C66024118","wikidata":"https://www.wikidata.org/wiki/Q1122506","display_name":"Computational model","level":2,"score":0.3522999882698059},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3391000032424927},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.3098999857902527},{"id":"https://openalex.org/C63882131","wikidata":"https://www.wikidata.org/wiki/Q17122954","display_name":"Strengths and weaknesses","level":2,"score":0.2865000069141388},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.28349998593330383},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.26759999990463257},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.2549000084400177}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.19307","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.19307","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.19307","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.19307","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Visual":[0],"Question":[1],"Answering":[2],"(VQA),":[3],"as":[4,10,156],"the":[5,16,144],"representative":[6],"multimodal":[7,116,163],"task,":[8],"serves":[9],"a":[11,55,157],"key":[12],"benchmark":[13],"for":[14,146],"evaluating":[15],"reasoning":[17],"capabilities":[18],"of":[19,75],"Multimodal":[20],"Large":[21],"Language":[22],"Models":[23],"(MLLMs).":[24],"However,":[25],"existing":[26],"evaluations":[27],"largely":[28],"rely":[29],"on":[30,79,108],"static":[31],"datasets":[32],"and":[33,42,82,112,152],"accuracy-based":[34],"metrics,":[35,130],"which":[36],"fail":[37],"to":[38,63,90,104],"capture":[39],"robustness,":[40],"consistency,":[41],"generalization.":[43],"Inspired":[44],"by":[45],"Metamorphic":[46,51,60],"Testing":[47],"(MT),":[48],"we":[49],"propose":[50],"Robustness":[52],"Assessment":[53],"(MetaRA),":[54],"testing":[56],"framework":[57],"that":[58,121,134],"employs":[59],"Relations":[61],"(MRs)":[62],"systematically":[64],"probe":[65],"vulnerabilities":[66],"in":[67,115,150],"MLLM-based":[68,92],"VQA":[69,93,151],"systems.":[70],"MetaRA":[71,89,122],"generates":[72],"controlled":[73],"variations":[74],"image-question":[76],"inputs":[77],"based":[78],"specific":[80],"MRs":[81],"evaluates":[83],"models":[84,94],"across":[85,95],"diverse":[86],"conditions.":[87],"Applying":[88],"multiple":[91],"different":[96],"tasks":[97],"reveals":[98],"nuanced":[99],"failure":[100,132],"patterns,":[101],"including":[102],"sensitivity":[103],"linguistic":[105],"perturbations,":[106],"over-reliance":[107],"superficial":[109],"visual":[110],"cues,":[111],"deeper":[113],"weaknesses":[114],"reasoning.":[117],"Experimental":[118],"results":[119],"demonstrate":[120],"provides":[123],"richer":[124],"diagnostic":[125],"insights":[126],"than":[127],"conventional":[128],"accuracy":[129],"exposing":[131],"modes":[133],"remain":[135],"hidden":[136],"under":[137],"standard":[138],"benchmarks.":[139],"Overall,":[140],"this":[141],"work":[142],"highlights":[143],"need":[145],"systematic":[147],"robustness":[148],"evaluation":[149],"positions":[153],"metamorphic":[154],"assessment":[155],"scalable,":[158],"model-agnostic":[159],"approach":[160],"toward":[161],"trustworthy":[162],"AI.":[164]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-21T00:00:00"}