{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,6]],"date-time":"2026-06-06T17:02:04Z","timestamp":1780765324352,"version":"3.54.1"},"reference-count":62,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"6","license":[{"start":{"date-parts":[[2024,6,1]],"date-time":"2024-06-01T00:00:00Z","timestamp":1717200000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2024,6,1]],"date-time":"2024-06-01T00:00:00Z","timestamp":1717200000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2024,6,1]],"date-time":"2024-06-01T00:00:00Z","timestamp":1717200000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Pattern Anal. Mach. Intell."],"published-print":{"date-parts":[[2024,6]]},"DOI":"10.1109\/tpami.2024.3355155","type":"journal-article","created":{"date-parts":[[2024,1,17]],"date-time":"2024-01-17T18:25:40Z","timestamp":1705515940000},"page":"4504-4518","source":"Crossref","is-referenced-by-count":23,"title":["B-Cos Alignment for Inherently Interpretable CNNs and Vision Transformers"],"prefix":"10.1109","volume":"46","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-5479-3769","authenticated-orcid":false,"given":"Moritz","family":"B\u00f6hle","sequence":"first","affiliation":[{"name":"Department of Computer Vision and Machine Learning, Max Planck Institute for Informatics, Saarland Informatics Campus, Saarbrücken, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2921-4876","authenticated-orcid":false,"given":"Navdeeppal","family":"Singh","sequence":"additional","affiliation":[{"name":"Department of Computer Vision and Machine Learning, Max Planck Institute for Informatics, Saarland Informatics Campus, Saarbrücken, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8949-9896","authenticated-orcid":false,"given":"Mario","family":"Fritz","sequence":"additional","affiliation":[{"name":"CISPA Helmholtz Center for Information Security, Saarbrücken, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9683-5237","authenticated-orcid":false,"given":"Bernt","family":"Schiele","sequence":"additional","affiliation":[{"name":"Department of Computer Vision and Machine Learning, Max Planck Institute for Informatics, Saarland Informatics Campus, Saarbrücken, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/JPROC.2021.3060483"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.5555\/3104322.3104425"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.5555\/2969033.2969153"},{"key":"ref4","article-title":"Benchmarking attribution methods with relative feature importance","author":"Yang","year":"2019"},{"key":"ref5","first-page":"3145","article-title":"Learning important features through propagating activation differences","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Shrikumar"},{"key":"ref6","first-page":"9525","article-title":"Sanity checks for saliency maps","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Adebayo"},{"key":"ref7","article-title":"Learning multiple layers of features from tiny images","author":"Krizhevsky","year":"2009"},{"key":"ref8","article-title":"An image is worth 16x16 words: Transformers for image recognition at scale","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Dosovitskiy"},{"key":"ref9","first-page":"1","article-title":"Very deep convolutional networks for large-scale image recognition","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Simonyan"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.243"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.634"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01167"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01008"},{"key":"ref15","first-page":"4768","article-title":"A unified approach to interpreting model predictions","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Lundberg"},{"key":"ref16","first-page":"151","article-title":"RISE: Randomized input sampling for explanation of black-box models","volume-title":"Proc. Brit. Mach. Vis. Conf.","author":"Petsiuk"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/N16-3020"},{"key":"ref18","first-page":"2668","article-title":"Interpretability beyond feature attribution: Quantitative testing with concept activation vectors (TCAV)","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Kim"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/ICDMW51313.2020.00068"},{"key":"ref20","article-title":"Deep inside convolutional networks: Visualising image classification models and saliency maps","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Simonyan"},{"key":"ref21","article-title":"Striving for simplicity: The all convolutional net","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Springenberg"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.319"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0130140"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.74"},{"key":"ref25","first-page":"4124","article-title":"Full-gradient representation for neural network visualization","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Srinivas"},{"key":"ref26","first-page":"3319","article-title":"Axiomatic attribution for deep networks","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Sundararajan"},{"key":"ref27","first-page":"3319","article-title":"This looks like that: Deep learning for interpretable image recognition","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Chen"},{"key":"ref28","article-title":"Approximating CNNs with bag-of-local-features models works surprisingly well on ImageNet","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Brendel"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00990"},{"key":"ref30","article-title":"Robustness may be at odds with accuracy","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Tsipras"},{"key":"ref31","article-title":"Do input gradients highlight discriminative features?","author":"Shah","year":"2021"},{"key":"ref32","article-title":"Bridging adversarial robustness and gradient interpretability","volume-title":"Proc. Safe Mach. Learn. Workshop Int. Conf. Learn. Representations","author":"Kim"},{"key":"ref33","article-title":"Rethinking the role of gradient-based attribution methods for model interpretability","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Srinivas"},{"key":"ref34","article-title":"Learning how to explain neural networks: PatternNet and PatternAttribution","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Kindermans"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.510"},{"key":"ref36","first-page":"3950","article-title":"Deep hyperspherical learning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Liu"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00293"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01418-6_38"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1007\/s11063-019-10043-7"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00012"},{"key":"ref41","first-page":"6659","article-title":"Bias also matters: Bias attribution for deep neural network explanation","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Wang"},{"key":"ref42","article-title":"Robust and interpretable blind image denoising via bias-free convolutional neural networks","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Mohan"},{"key":"ref43","first-page":"19513","article-title":"Fast axiomatic attribution for neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Hesse"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P19-1282"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.acl-main.385"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.blackboxnlp-1.14"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1145\/3459637.3482126"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00084"},{"key":"ref50","article-title":"Searching for activation functions","volume-title":"Proc. Int. Conf. Learn. Representations Workshop","author":"Ramachandran"},{"key":"ref51","first-page":"319","article-title":"Maxout networks","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Goodfellow"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.5555\/3045118.3045167"},{"key":"ref53","article-title":"Layer Normalization","author":"Ba","year":"2016"},{"key":"ref54","first-page":"1622","article-title":"Positional normalization","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Li"},{"key":"ref55","article-title":"Instance normalization: The missing ingredient for fast stylization","author":"Ulyanov","year":"2016"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01261-8_1"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref58","first-page":"8026","article-title":"PyTorch: An imperative style, high-performance deep learning library","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Paszke"},{"key":"ref59","article-title":"Better plain ViT baselines for ImageNet-1k","author":"Beyer","year":"2022"},{"key":"ref60","article-title":"Early convolutions help transformers see better","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Xiao"},{"key":"ref63","first-page":"1803","article-title":"How to explain individual classification decisions","author":"Baehrens","year":"2010","journal-title":"J. Mach. Learn. Res."},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00998"}],"container-title":["IEEE Transactions on Pattern Analysis and Machine Intelligence"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/34\/10522060\/10401936.pdf?arnumber=10401936","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,5,8]],"date-time":"2024-05-08T04:42:21Z","timestamp":1715143341000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10401936\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,6]]},"references-count":62,"journal-issue":{"issue":"6"},"URL":"https:\/\/doi.org\/10.1109\/tpami.2024.3355155","relation":{},"ISSN":["0162-8828","2160-9292","1939-3539"],"issn-type":[{"value":"0162-8828","type":"print"},{"value":"2160-9292","type":"electronic"},{"value":"1939-3539","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,6]]}}}