{"id":"https://openalex.org/W7137839915","doi":"https://doi.org/10.1609/aaai.v40i9.37620","title":"Anatomical Region-Guided Contrastive Decoding: A Plug-and-Play Strategy for Mitigating Hallucinations in Medical VLMs","display_name":"Anatomical Region-Guided Contrastive Decoding: A Plug-and-Play Strategy for Mitigating Hallucinations in Medical VLMs","publication_year":2026,"publication_date":"2026-03-14","ids":{"openalex":"https://openalex.org/W7137839915","doi":"https://doi.org/10.1609/aaai.v40i9.37620"},"language":null,"primary_location":{"id":"doi:10.1609/aaai.v40i9.37620","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i9.37620","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://doi.org/10.1609/aaai.v40i9.37620","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5129687771","display_name":"Xiao Liang","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Xiao Liang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129741762","display_name":"Chenxi Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chenxi Liu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5121118821","display_name":"Zhi Ma","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhi Ma","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129671181","display_name":"Di Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Di Wang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129707378","display_name":"Bin Jing","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bin Jing","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129714788","display_name":"Quan Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Quan Wang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5129729487","display_name":"Yuanyuan Shi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuanyuan Shi","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5129687771"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.05965697,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"40","issue":"9","first_page":"6871","last_page":"6879"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.30730000138282776,"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":0.30730000138282776,"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/T11775","display_name":"COVID-19 diagnosis using AI","score":0.2687000036239624,"subfield":{"id":"https://openalex.org/subfields/2741","display_name":"Radiology, Nuclear Medicine and Imaging"},"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":0.0551999993622303,"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/decoding-methods","display_name":"Decoding methods","score":0.5701000094413757},{"id":"https://openalex.org/keywords/limiting","display_name":"Limiting","score":0.5584999918937683},{"id":"https://openalex.org/keywords/reliability","display_name":"Reliability (semiconductor)","score":0.512499988079071},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.42179998755455017},{"id":"https://openalex.org/keywords/contrast","display_name":"Contrast (vision)","score":0.28060001134872437},{"id":"https://openalex.org/keywords/psychological-intervention","display_name":"Psychological intervention","score":0.2727000117301941}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6266999840736389},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.5701000094413757},{"id":"https://openalex.org/C188198153","wikidata":"https://www.wikidata.org/wiki/Q1613840","display_name":"Limiting","level":2,"score":0.5584999918937683},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.512499988079071},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.42179998755455017},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.4156000018119812},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.41370001435279846},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.373199999332428},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.28060001134872437},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2782999873161316},{"id":"https://openalex.org/C27415008","wikidata":"https://www.wikidata.org/wiki/Q7256382","display_name":"Psychological intervention","level":2,"score":0.2727000117301941},{"id":"https://openalex.org/C2780910867","wikidata":"https://www.wikidata.org/wiki/Q1952416","display_name":"Multimodality","level":2,"score":0.266400009393692},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.26570001244544983},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.2621999979019165},{"id":"https://openalex.org/C2779974597","wikidata":"https://www.wikidata.org/wiki/Q28448986","display_name":"Clinical Practice","level":2,"score":0.258899986743927},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.258899986743927},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2502000033855438}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1609/aaai.v40i9.37620","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i9.37620","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1609/aaai.v40i9.37620","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i9.37620","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"sustainable_development_goals":[{"score":0.4479646682739258,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Medical":[0],"Vision-Language":[1],"Models":[2],"(MedVLMs)":[3],"show":[4],"immense":[5],"promise":[6],"in":[7,65,153],"clinical":[8,68],"applicability.":[9],"However,":[10],"their":[11],"reliability":[12],"is":[13],"hindered":[14],"by":[15,90],"hallucinations,":[16,158],"where":[17],"models":[18],"often":[19],"fail":[20],"to":[21,101],"derive":[22],"answers":[23],"from":[24],"visual":[25],"evidence,":[26],"instead":[27],"relying":[28],"on":[29,44],"learned":[30],"textual":[31],"priors.":[32],"Existing":[33],"mitigation":[34],"strategies":[35],"for":[36],"MedVLMs":[37],"have":[38],"distinct":[39],"limitations:":[40],"training-based":[41],"methods":[42],"rely":[43],"costly":[45],"expert":[46],"annotations,":[47],"limiting":[48],"scalability,":[49],"while":[50],"training-free":[51],"interventions":[52],"like":[53],"contrastive":[54,105],"decoding,":[55],"though":[56],"data-efficient,":[57],"apply":[58],"a":[59,84,103],"global,":[60],"untargeted":[61],"correction":[62],"whose":[63],"effects":[64],"complex":[66],"real-world":[67],"settings":[69],"can":[70],"be":[71],"unreliable.":[72],"To":[73],"address":[74],"these":[75],"challenges,":[76],"we":[77],"introduce":[78],"Anatomical":[79],"Region-Guided":[80],"Contrastive":[81],"Decoding":[82],"(ARCD),":[83],"plug-and-play":[85],"strategy":[86],"that":[87],"mitigates":[88],"hallucinations":[89],"providing":[91],"targeted,":[92],"region-specific":[93],"guidance.":[94],"Our":[95],"module":[96],"leverages":[97],"an":[98],"anatomical":[99,128],"mask":[100],"direct":[102],"three-tiered":[104],"decoding":[106],"process.":[107],"By":[108],"dynamically":[109],"re-weighting":[110],"at":[111],"the":[112,121],"token,":[113],"attention,":[114],"and":[115,130,146,159],"logits":[116],"levels,":[117],"it":[118],"verifiably":[119],"steers":[120],"model's":[122],"focus":[123],"onto":[124],"specified":[125],"regions,":[126],"reinforcing":[127],"understanding":[129],"suppressing":[131],"factually":[132],"incorrect":[133],"outputs.":[134],"Extensive":[135],"experiments":[136],"across":[137],"diverse":[138],"datasets,":[139],"including":[140],"chest":[141],"X-ray,":[142],"CT,":[143],"brain":[144],"MRI,":[145],"ocular":[147],"ultrasound,":[148],"demonstrate":[149],"our":[150],"method's":[151],"effectiveness":[152],"improving":[154],"regional":[155],"understanding,":[156],"reducing":[157],"enhancing":[160],"overall":[161],"diagnostic":[162],"accuracy.":[163]},"counts_by_year":[],"updated_date":"2026-03-18T06:31:55.123368","created_date":"2026-03-18T00:00:00"}