{"id":"https://openalex.org/W7155358239","doi":"https://doi.org/10.48448/54wp-8f04","title":"Benchmarking and Mitigating the Impact of Noisy User Prompts in Medical VLMs via Cross-Modal Reflection","display_name":"Benchmarking and Mitigating the Impact of Noisy User Prompts in Medical VLMs via Cross-Modal Reflection","publication_year":2026,"publication_date":"2026-03-06","ids":{"openalex":"https://openalex.org/W7155358239","doi":"https://doi.org/10.48448/54wp-8f04"},"language":"en","primary_location":{"id":"doi:10.48448/54wp-8f04","doi":"https://doi.org/10.48448/54wp-8f04","is_oa":true,"landing_page_url":"https://doi.org/10.48448/54wp-8f04","pdf_url":null,"source":{"id":"https://openalex.org/S7407053148","display_name":"Underline Science Inc.","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Audiovisual"},"volume":null,"type":"other","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48448/54wp-8f04","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5134424196","display_name":"Association for Computational Linguistics 2026","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Association for Computational Linguistics 2026","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029714016","display_name":"Reza Abbasi-Asl","orcid":"https://orcid.org/0000-0001-7824-4628","observed_orcids":["https://orcid.org/0000-0001-7824-4628"]},"institutions":[{"id":"https://openalex.org/I180670191","display_name":"University of California, San Francisco","ror":"https://ror.org/043mz5j54","country_code":"US","type":"education","lineage":["https://openalex.org/I180670191"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Abbasi-Asl, Reza","raw_affiliation_strings":["University of California, San Francisco"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of California, San Francisco","institution_ids":["https://openalex.org/I180670191"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040270189","display_name":"Ramtin Pedarsani","orcid":"https://orcid.org/0000-0002-1126-0292","observed_orcids":["https://orcid.org/0000-0002-1126-0292","https://orcid.org/0000-0003-0786-2132"]},"institutions":[{"id":"https://openalex.org/I154570441","display_name":"University of California, Santa Barbara","ror":"https://ror.org/02t274463","country_code":"US","type":"education","lineage":["https://openalex.org/I154570441"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Pedarsani, Ramtin","raw_affiliation_strings":["UC Santa Barbara"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Santa Barbara","institution_ids":["https://openalex.org/I154570441"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5040852802","display_name":"Zhiyu Xue","orcid":"https://orcid.org/0009-0009-1389-0499","observed_orcids":["https://orcid.org/0009-0009-1389-0499"]},"institutions":[{"id":"https://openalex.org/I154570441","display_name":"University of California, Santa Barbara","ror":"https://ror.org/02t274463","country_code":"US","type":"education","lineage":["https://openalex.org/I154570441"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xue, Zhiyu","raw_affiliation_strings":["UCSB"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UCSB","institution_ids":["https://openalex.org/I154570441"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"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":0.9980000257492065,"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":0.9980000257492065,"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/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.0017000000225380063,"subfield":{"id":"https://openalex.org/subfields/2718","display_name":"Health Informatics"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.0,"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/medical-vision-language-models","display_name":"medical vision-language models","score":0.9959999918937683},{"id":"https://openalex.org/keywords/prompt-robustness","display_name":"prompt robustness","score":0.859000027179718},{"id":"https://openalex.org/keywords/chain-of-thought-fine-tuning","display_name":"chain-of-thought fine-tuning","score":0.8539999723434448},{"id":"https://openalex.org/keywords/medical-question-answering","display_name":"medical question answering","score":0.6380000114440918},{"id":"https://openalex.org/keywords/medical-image-diagnosis","display_name":"medical image diagnosis","score":0.47200000286102295},{"id":"https://openalex.org/keywords/reasoning-paths","display_name":"reasoning paths","score":0.5099999904632568},{"id":"https://openalex.org/keywords/out-of-domain-evaluation","display_name":"out-of-domain evaluation","score":0.6909999847412109}],"concepts":[{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.7250999808311462},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.6953999996185303},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6814000010490417},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5601000189781189},{"id":"https://openalex.org/C65682993","wikidata":"https://www.wikidata.org/wiki/Q1056451","display_name":"Reflection (computer programming)","level":2,"score":0.5389999747276306},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.49559998512268066},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4724000096321106},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.42890000343322754},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.42640000581741333},{"id":"https://openalex.org/C534262118","wikidata":"https://www.wikidata.org/wiki/Q177719","display_name":"Medical diagnosis","level":2,"score":0.367900013923645},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.3621000051498413},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.3619000017642975},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.3334999978542328},{"id":"https://openalex.org/C31601959","wikidata":"https://www.wikidata.org/wiki/Q931309","display_name":"Medical imaging","level":2,"score":0.2955999970436096},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.2897999882698059},{"id":"https://openalex.org/C138268822","wikidata":"https://www.wikidata.org/wiki/Q1051925","display_name":"Resolution (logic)","level":2,"score":0.273499995470047},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.27070000767707825},{"id":"https://openalex.org/C106977388","wikidata":"https://www.wikidata.org/wiki/Q2752427","display_name":"Medical research","level":2,"score":0.2563999891281128}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48448/54wp-8f04","doi":"https://doi.org/10.48448/54wp-8f04","is_oa":true,"landing_page_url":"https://doi.org/10.48448/54wp-8f04","pdf_url":null,"source":{"id":"https://openalex.org/S7407053148","display_name":"Underline Science Inc.","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Audiovisual"}],"best_oa_location":{"id":"doi:10.48448/54wp-8f04","doi":"https://doi.org/10.48448/54wp-8f04","is_oa":true,"landing_page_url":"https://doi.org/10.48448/54wp-8f04","pdf_url":null,"source":{"id":"https://openalex.org/S7407053148","display_name":"Underline Science Inc.","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Audiovisual"},"sustainable_development_goals":[{"display_name":"Good health and well-being","id":"https://metadata.un.org/sdg/3","score":0.5683000087738037}],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4412528379","https://openalex.org/W7154221963","https://openalex.org/W7155212901","https://openalex.org/W7106209332","https://openalex.org/W4408492863","https://openalex.org/W7161677896","https://openalex.org/W7211990266","https://openalex.org/W7127443799","https://openalex.org/W7140309924","https://openalex.org/W7077874608"],"abstract_inverted_index":{"Medical":[0],"vision-language":[1],"models":[2],"(Med-VLMs)":[3],"offer":[4],"a":[5,50,56,114,164],"new":[6],"and":[7,22,37,77,131,152,169,189],"effective":[8],"paradigm":[9],"for":[10,120,145,186],"digital":[11],"health":[12],"in":[13,105],"tasks":[14],"such":[15],"as":[16],"disease":[17],"diagnosis":[18],"using":[19],"clinical":[20,41,70],"images":[21,130],"text.":[23,132],"In":[24,133],"these":[25],"tasks,":[26],"an":[27],"important":[28],"but":[29],"underexplored":[30],"research":[31],"question":[32,59],"is":[33,139,160],"how":[34],"Med-VLMs":[35,85,121],"interpret":[36],"respond":[38],"to":[39,64,87,141],"user-provided":[40,89,154,184],"information,":[42],"especially":[43],"when":[44],"the":[45,67,137,146,149,153,157,167,180],"prompts":[46,71,90,185],"are":[47,96],"noisy.":[48],"For":[49],"systematic":[51],"evaluation,":[52],"we":[53,112],"construct":[54],"Med-CP,":[55],"large-scale":[57],"visual":[58,168],"answering":[60],"(VQA)":[61],"benchmark":[62],"designed":[63],"comprehensively":[65],"evaluate":[66],"influence":[68],"of":[69,93,148],"across":[72,128],"diverse":[73],"modalities,":[74],"anatomical":[75],"regions,":[76],"diagnostic":[78],"tasks.":[79],"Our":[80],"experiments":[81],"reveal":[82],"that":[83,175],"existing":[84],"tend":[86],"follow":[88],"blindly,":[91],"regardless":[92],"whether":[94],"they":[95],"accurate":[97],"or":[98],"not,":[99],"raising":[100],"concerns":[101],"about":[102],"their":[103],"reliability":[104],"real-world":[106],"interactions.":[107],"To":[108],"address":[109],"this":[110],"problem,":[111],"introduce":[113],"novel":[115],"supervised":[116],"fine-tuning":[117],"(SFT)":[118],"approach":[119],"based":[122],"on":[123,166],"cross-modal":[124],"reflection":[125,165],"chain-of-thought":[126],"(CoT)":[127],"medical":[129,150],"our":[134,176],"SFT":[135],"method,":[136],"Med-VLM":[138],"trained":[140],"produce":[142],"reasoning":[143],"paths":[144],"analysis":[147],"image":[151],"prompt.":[155],"Then,":[156],"final":[158],"answer":[159],"determined":[161],"by":[162],"conducting":[163],"textual":[170],"information.":[171],"Experimental":[172],"results":[173],"demonstrate":[174],"method":[177],"considerably":[178],"enhances":[179],"robustness":[181],"against":[182],"noisy":[183],"both":[187],"in-domain":[188],"out-of-domain":[190],"evaluation":[191],"scenarios.":[192]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-04-24T00:00:00"}