{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,6]],"date-time":"2026-05-06T15:47:57Z","timestamp":1778082477268,"version":"3.51.4"},"reference-count":72,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,10,19]],"date-time":"2025-10-19T00:00:00Z","timestamp":1760832000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,10,19]],"date-time":"2025-10-19T00:00:00Z","timestamp":1760832000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/100010663","name":"ERC","doi-asserted-by":"publisher","award":["853489DEXIM"],"award-info":[{"award-number":["853489DEXIM"]}],"id":[{"id":"10.13039\/100010663","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100005306","name":"Alfried Krupp von Bohlen und Halbach Foundation","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100005306","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100019692","name":"Leibniz Supercomputing Centre (LRZ)","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100019692","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,10,19]]},"DOI":"10.1109\/iccv51701.2025.02153","type":"proceedings-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:45:49Z","timestamp":1777491949000},"page":"23188-23198","source":"Crossref","is-referenced-by-count":1,"title":["SUB: Benchmarking CBM Generalization via Synthetic Attribute Substitutions"],"prefix":"10.1109","author":[{"given":"Jessica","family":"Bader","sequence":"first","affiliation":[{"name":"Technical University of Munich,Helmholtz Munich, Munich Center for Machine Learning (MCML)"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Leander","family":"Girrbach","sequence":"additional","affiliation":[{"name":"Technical University of Munich,Helmholtz Munich, Munich Center for Machine Learning (MCML)"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Stephan","family":"Alaniz","sequence":"additional","affiliation":[{"name":"Institut Polytechnique de Paris,LTCI, Télécom Paris,France"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zeynep","family":"Akata","sequence":"additional","affiliation":[{"name":"Technical University of Munich,Helmholtz Munich, Munich Center for Machine Learning (MCML)"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Towards interpretable radiology report generation via concept bottlenecks using a multiagentic rag","author":"Md","year":"2024","journal-title":"arXiv"},{"key":"ref2","article-title":"Towards robust interpretability with self-explaining neural networks","author":"Alvarez-Melis","year":"2018","journal-title":"NeurIPS"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73013-9_15"},{"key":"ref4","article-title":"Sega: Instructing text-to-image models using semantic guidance","author":"Brack","year":"2024","journal-title":"NeurIPS"},{"key":"ref5","article-title":"This looks like that: deep learning for interpretable image recognition","author":"Chen","year":"2018","journal-title":"NeurIPS"},{"key":"ref6","article-title":"Enhancing prompt following with visual control through training-free mask-guided diffusion","author":"Chen","year":"2024","journal-title":"arXiv"},{"key":"ref7","article-title":"Expanding performance boundaries of open-source multimodal models with model, data, and testtime scaling","author":"Chen","year":"2024","journal-title":"arXiv"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1007\/s11432-024-4231-5"},{"key":"ref9","article-title":"Internvl: Scaling up vision foundation models and aligning for generic visual-linguistic tasks","author":"Chen","year":"2024","journal-title":"CVPR"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72117-5_4"},{"key":"ref11","article-title":"Sparse autoencoders find highly interpretable features in language models","author":"Cunningham","year":"2023","journal-title":"arXiv"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.316"},{"key":"ref13","article-title":"Position: Compositional generative modeling: A single model is not all you need","author":"Du","year":"2024","journal-title":"ICML"},{"key":"ref14","article-title":"Compositional visual generation with energy based models","author":"Du","year":"2020","journal-title":"NeurIPS"},{"key":"ref15","article-title":"Reduce, reuse, recycle:Compositional generation with energy-based diffusion models and mcmc","volume-title":"ICML","author":"Du","year":"2023"},{"key":"ref16","article-title":"Diversify your vision datasets with automatic diffusion-based augmentation","author":"Dunlap","year":"2023","journal-title":"NeurIPS"},{"key":"ref17","article-title":"Reno: Enhancing one-step text-to-image models through reward-based noise optimization","author":"Vincent Eyring","year":"2024","journal-title":"NeurIPS"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00454"},{"key":"ref19","author":"Furby","year":"2024","journal-title":"Can we constrain concept bottleneck models to learn semantically meaningful input features?"},{"key":"ref20","article-title":"Coind: Enabling logical compositions in diffusion models","author":"Gaudi","year":"2025","journal-title":"ICLR"},{"key":"ref21","article-title":"Addressing leakage in concept bottleneck models","author":"Havasi","year":"2022","journal-title":"NeurIPS"},{"key":"ref22","article-title":"Is synthetic data from generative models ready for image recognition?","author":"He","year":"2023","journal-title":"ICLR"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/WACV56688.2023.00476"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00368"},{"key":"ref25","article-title":"Classifier-free diffusion guidance","author":"Ho","year":"2021","journal-title":"NeurIPS Workshop on Deep Generative Models and Downstream Applications"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298658"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.215"},{"key":"ref28","article-title":"Revelio: Interpreting and leveraging semantic information in diffusion models","author":"Kim","year":"2024","journal-title":"arXiv"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73209-6_15"},{"key":"ref30","article-title":"Concept bottleneck models","author":"Wei Koh","year":"2020","journal-title":"ICML"},{"key":"ref31","article-title":"Announcing black forest labs","year":"2024","journal-title":"BlackForestLabs Blog"},{"key":"ref32","article-title":"Learning to compose visual relations","author":"Liu","year":"2021","journal-title":"NeurIPS"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19790-1_26"},{"key":"ref34","article-title":"Promises and pitfalls of black-box concept learning models","author":"Mahinpei","year":"2021","journal-title":"ArXiv"},{"key":"ref35","article-title":"Frey. k-sparse autoencoders","author":"Makhzani","year":"2013","journal-title":"arXiv"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1542"},{"key":"ref37","article-title":"Do concept bottleneck models learn as intended?","author":"Margeloiu","year":"2021","journal-title":"arXiv"},{"key":"ref38","article-title":"Visual classification via description from large language models","author":"Menon","year":"2023","journal-title":"ICLR"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00713"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00269"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/BHI58575.2023.10313382"},{"key":"ref42","article-title":"Label-free concept bottleneck models","author":"Oikarinen","year":"2023","journal-title":"ICLR"},{"key":"ref43","article-title":"Lucidppn: Unambiguous prototypical parts network for user-centric interpretable computer vision","author":"Pach","year":"2024","journal-title":"arXiv"},{"key":"ref44","article-title":"Coarse-to-fine concept bottleneck models","author":"Panousis","year":"2024","journal-title":"NeurIPS"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00472"},{"key":"ref46","article-title":"Sdxl: Improving latent diffusion models for high-resolution image synthesis","volume-title":"ICLR","author":"Podell","year":"2024"},{"key":"ref47","article-title":"Learning transferable visual models from natural language supervision","author":"Radford","year":"2021","journal-title":"ICML"},{"key":"ref48","author":"Raman","year":"2024","journal-title":"Do concept bottleneck models respect localities?"},{"key":"ref49","article-title":"Understanding inter-concept relationships in conceptbased models","author":"Raman","year":"2024","journal-title":"ICML"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72980-5_26"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01042"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1145\/3447548.3467245"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19775-8_21"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00774"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2022.3167702"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73347-5_24"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v37i12.26765"},{"key":"ref58","article-title":"Eva-clip: Improved training techniques for clip at scale","author":"Sun","year":"2023","journal-title":"arXiv"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73016-0_8"},{"key":"ref60","article-title":"Universal sparse autoencoders: Interpretable cross-model concept alignment","author":"Thasarathan","year":"2025","journal-title":"arXiv"},{"key":"ref61","article-title":"Siglip 2: Multilingual vision-language encoders with improved semantic understanding, localization, and dense features","author":"Tschannen","year":"2025","journal-title":"arXiv"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00179"},{"key":"ref63","article-title":"The caltech-ucsd birds-200\u20132011 dataset","author":"Wah","year":"2011","journal-title":"California Institute of Technology Technical Report"},{"key":"ref64","article-title":"On discrete prompt optimization for diffusion models","author":"Wang","year":"2024","journal-title":"ICML"},{"key":"ref65","article-title":"Compositional generalization from first principles","author":"Wiedemer","year":"2023","journal-title":"NeurIPS"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01839"},{"key":"ref67","article-title":"Concept embedding models","author":"Espinosa Zarlenga","year":"2022","journal-title":"NeurIPS"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v37i10.26392"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01100"},{"key":"ref70","article-title":"Large multi-modal models can interpret features in large multimodal models","author":"Zhang","year":"2024","journal-title":"arXiv"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00355"},{"key":"ref72","article-title":"Uni-controlnet: All-in-one control to text-to-image diffusion models","author":"Zhao","year":"2023","journal-title":"NeurIPS"}],"event":{"name":"2025 IEEE\/CVF International Conference on Computer Vision (ICCV)","location":"Honolulu, HI, USA","start":{"date-parts":[[2025,10,19]]},"end":{"date-parts":[[2025,10,25]]}},"container-title":["2025 IEEE\/CVF International Conference on Computer Vision (ICCV)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11443115\/11443287\/11444626.pdf?arnumber=11444626","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T06:46:17Z","timestamp":1777531577000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11444626\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":72,"URL":"https:\/\/doi.org\/10.1109\/iccv51701.2025.02153","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}