{"id":"https://openalex.org/W7161755193","doi":"https://doi.org/10.1109/isbi61048.2026.11515430","title":"Measuring and Aligning Abstraction in Vision-Language Models with Medical Taxonomies","display_name":"Measuring and Aligning Abstraction in Vision-Language Models with Medical Taxonomies","publication_year":2026,"publication_date":"2026-04-08","ids":{"openalex":"https://openalex.org/W7161755193","doi":"https://doi.org/10.1109/isbi61048.2026.11515430"},"language":null,"primary_location":{"id":"doi:10.1109/isbi61048.2026.11515430","is_oa":false,"landing_page_url":"https://doi.org/10.1109/isbi61048.2026.11515430","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE 23rd International Symposium on Biomedical Imaging (ISBI)","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/A5080375784","display_name":"Ben Schaper","orcid":null},"institutions":[{"id":"https://openalex.org/I62916508","display_name":"Technical University of Munich","ror":"https://ror.org/02kkvpp62","country_code":"DE","type":"education","lineage":["https://openalex.org/I62916508"]}],"countries":["DE"],"is_corresponding":true,"raw_author_name":"Ben Schaper","raw_affiliation_strings":["School of Computation, Information and Technology, Technical University of Munich,Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computation, Information and Technology, Technical University of Munich,Germany","institution_ids":["https://openalex.org/I62916508"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5067896424","display_name":"Maxime Di Folco","orcid":"https://orcid.org/0000-0001-6160-8050"},"institutions":[{"id":"https://openalex.org/I62916508","display_name":"Technical University of Munich","ror":"https://ror.org/02kkvpp62","country_code":"DE","type":"education","lineage":["https://openalex.org/I62916508"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Maxime Di Folco","raw_affiliation_strings":["School of Computation, Information and Technology, Technical University of Munich,Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computation, Information and Technology, Technical University of Munich,Germany","institution_ids":["https://openalex.org/I62916508"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136505839","display_name":"Bernhard Kainz","orcid":null},"institutions":[{"id":"https://openalex.org/I4210164862","display_name":"Artificial Intelligence in Medicine (Canada)","ror":"https://ror.org/05p590m36","country_code":"CA","type":"company","lineage":["https://openalex.org/I4210164862"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Bernhard Kainz","raw_affiliation_strings":["FAU Erlangen-Nuremberg,Department of Artificial Intelligence in Biomedical Imaging,Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"FAU Erlangen-Nuremberg,Department of Artificial Intelligence in Biomedical Imaging,Germany","institution_ids":["https://openalex.org/I4210164862"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136581304","display_name":"Julia A. Schnabel","orcid":null},"institutions":[{"id":"https://openalex.org/I62916508","display_name":"Technical University of Munich","ror":"https://ror.org/02kkvpp62","country_code":"DE","type":"education","lineage":["https://openalex.org/I62916508"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Julia A. Schnabel","raw_affiliation_strings":["School of Computation, Information and Technology, Technical University of Munich,Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computation, Information and Technology, Technical University of Munich,Germany","institution_ids":["https://openalex.org/I62916508"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5086799838","display_name":"Cosmin I. Bercea","orcid":"https://orcid.org/0000-0003-2628-2766"},"institutions":[{"id":"https://openalex.org/I62916508","display_name":"Technical University of Munich","ror":"https://ror.org/02kkvpp62","country_code":"DE","type":"education","lineage":["https://openalex.org/I62916508"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Cosmin I. Bercea","raw_affiliation_strings":["School of Computation, Information and Technology, Technical University of Munich,Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computation, Information and Technology, Technical University of Munich,Germany","institution_ids":["https://openalex.org/I62916508"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5080375784"],"corresponding_institution_ids":["https://openalex.org/I62916508"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.92909575,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"4"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.6931999921798706,"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":0.6931999921798706,"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/T13702","display_name":"Machine Learning in Healthcare","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"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.05429999902844429,"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/abstraction","display_name":"Abstraction","score":0.5932000279426575},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.258899986743927},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.2567000091075897},{"id":"https://openalex.org/keywords/information-system","display_name":"Information system","score":0.2551000118255615}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6870999932289124},{"id":"https://openalex.org/C124304363","wikidata":"https://www.wikidata.org/wiki/Q673661","display_name":"Abstraction","level":2,"score":0.5932000279426575},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.38280001282691956},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.30219998955726624},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.28999999165534973},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.26089999079704285},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.258899986743927},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2567000091075897},{"id":"https://openalex.org/C180198813","wikidata":"https://www.wikidata.org/wiki/Q121182","display_name":"Information system","level":2,"score":0.2551000118255615},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.2517000138759613},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.2500999867916107}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/isbi61048.2026.11515430","is_oa":false,"landing_page_url":"https://doi.org/10.1109/isbi61048.2026.11515430","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE 23rd International Symposium on Biomedical Imaging (ISBI)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":8,"referenced_works":["https://openalex.org/W2115610681","https://openalex.org/W2332766554","https://openalex.org/W2801092024","https://openalex.org/W4385573131","https://openalex.org/W4386566421","https://openalex.org/W4390873312","https://openalex.org/W4404820070","https://openalex.org/W4405669265"],"related_works":[],"abstract_inverted_index":{"Vision-Language":[0],"Models":[1],"(VLMs)":[2],"show":[3],"strong":[4],"zero-shot":[5],"performance":[6],"for":[7,105],"chest":[8],"X-ray":[9],"classification,":[10],"but":[11],"standard":[12],"flat":[13,67],"metrics":[14,45],"fail":[15],"to":[16,28,51,87],"distinguish":[17],"between":[18],"clinically":[19,109],"minor":[20],"and":[21,30,46,76,102,107],"severe":[22,84],"errors.":[23],"This":[24],"work":[25],"investigates":[26],"how":[27],"quantify":[29],"mitigate":[31],"abstraction":[32,85],"errors":[33,86],"by":[34],"leveraging":[35],"medical":[36],"taxonomies.":[37],"We":[38],"benchmark":[39],"several":[40],"state-of-the-art":[41],"VLMs":[42,61],"using":[43],"hierarchical":[44,100],"introduce":[47],"Catastrophic":[48],"Abstraction":[49],"Errors":[50],"capture":[52],"cross-branch":[53],"mistakes.":[54],"Our":[55],"results":[56],"reveal":[57],"substantial":[58],"misalignment":[59],"of":[60,99,112],"with":[62,79],"clinical":[63],"taxonomies":[64],"despite":[65],"high":[66],"performance.":[68,93],"To":[69],"address":[70],"this,":[71],"we":[72],"propose":[73],"risk-constrained":[74],"thresholding":[75],"taxonomy-aware":[77],"fine-tuning":[78],"radial":[80],"embeddings,":[81],"which":[82],"reduce":[83],"below":[88],"2%":[89],"while":[90],"maintaining":[91],"competitive":[92],"These":[94],"findings":[95],"highlight":[96],"the":[97],"importance":[98],"evaluation":[101],"representation-level":[103],"alignment":[104],"safer":[106],"more":[108],"meaningful":[110],"deployment":[111],"VLMs.":[113]},"counts_by_year":[],"updated_date":"2026-05-22T06:13:13.366637","created_date":"2026-05-21T00:00:00"}