{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,17]],"date-time":"2026-11-17T12:17:16Z","timestamp":1794917836284,"version":"4.9.1"},"publisher-location":"California","reference-count":0,"publisher":"International Joint Conferences on Artificial Intelligence Organization","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,9]]},"abstract":"<jats:p>Vision-language model (VLM) encoders such as CLIP enable strong retrieval and zero-shot classification in a shared image\u2013text embedding space, yet the semantic organization of this space is rarely inspected. We present a post-hoc framework to explain, verify, and align the semantic hierarchies induced by a VLM over a given set of child classes. First, we extract a binary hierarchy by agglomerative clustering of class centroids and name internal nodes by dictionary-based matching to a concept bank. Second, we quantify plausibility by comparing the extracted tree against human ontologies using efficient tree- and edge-level consistency measures, and we evaluate utility via explainable hierarchical tree-traversal inference with uncertainty-aware early stopping (UAES). Third, we propose an ontology-guided post-hoc alignment method that learns a lightweight embedding-space transformation, using UMAP to generate target neighborhoods from a desired hierarchy. Across 13 pretrained VLMs and 4 image datasets, our method finds systematic modality differences: image encoders are more discriminative, while text encoders induce hierarchies that better match human taxonomies. Overall, the results reveal a persistent trade-off between zero-shot accuracy and ontological plausibility and suggest practical routes to improve semantic alignment in shared embedding spaces.<\/jats:p>","DOI":"10.24963\/ijcai.2026\/82","type":"proceedings-article","created":{"date-parts":[[2026,9,16]],"date-time":"2026-09-16T22:01:13Z","timestamp":1789596073000},"page":"724-734","source":"Crossref","is-referenced-by-count":0,"title":["Explaining, Verifying and Aligning Semantic Hierarchies in Vision-Language Model Embeddings"],"prefix":"10.24963","author":[{"given":"Gesina","family":"Schwalbe","sequence":"first","affiliation":[{"name":"Ulm University, Germany"},{"name":"University of L\u00fcbeck, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mert","family":"Keser","sequence":"additional","affiliation":[{"name":"Technical University of Munich, Germany"},{"name":"AUMOVIO SE, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Moritz","family":"Bayerkuhnlein","sequence":"additional","affiliation":[{"name":"University of L\u00fcbeck, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Edgar","family":"Heinert","sequence":"additional","affiliation":[{"name":"Osnabr\u00fcck University, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Annika","family":"M\u00fctze","sequence":"additional","affiliation":[{"name":"Osnabr\u00fcck University, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Marvin","family":"Keller","sequence":"additional","affiliation":[{"name":"University of L\u00fcbeck, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sparsh","family":"Tiwari","sequence":"additional","affiliation":[{"name":"University of L\u00fcbeck, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Georgii","family":"Mikriukov","sequence":"additional","affiliation":[{"name":"Leibniz Institute for Agricultural Engineering and Bioeconomy, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Diedrich","family":"Wolter","sequence":"additional","affiliation":[{"name":"University of L\u00fcbeck, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jae Hee","family":"Lee","sequence":"additional","affiliation":[{"name":"University of Hamburg, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Matthias","family":"Rottmann","sequence":"additional","affiliation":[{"name":"Osnabr\u00fcck University, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"10584","event":{"name":"Thirty-Fifth International Joint Conference on Artificial Intelligence {IJCAI-26}","theme":"Artificial Intelligence","location":"Bremen, Germany","acronym":"IJCAI-2026","number":"35","sponsor":["International Joint Conferences on Artificial Intelligence Organization (IJCAI)"],"start":{"date-parts":[[2026,8,15]]},"end":{"date-parts":[[2026,8,21]]}},"container-title":["Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"],"original-title":[],"deposited":{"date-parts":[[2026,11,17]],"date-time":"2026-11-17T11:27:43Z","timestamp":1794914863000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.ijcai.org\/proceedings\/2026\/82"}},"subtitle":[],"proceedings-subject":"Artificial Intelligence Research Articles","short-title":[],"issued":{"date-parts":[[2026,9]]},"references-count":0,"URL":"https:\/\/doi.org\/10.24963\/ijcai.2026\/82","relation":{},"subject":[],"published":{"date-parts":[[2026,9]]}}}