{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,3]],"date-time":"2026-07-03T17:54:26Z","timestamp":1783101266796,"version":"3.54.6"},"reference-count":42,"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"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,10,19]]},"DOI":"10.1109\/iccv51701.2025.01891","type":"proceedings-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:45:49Z","timestamp":1777491949000},"page":"1-10","source":"Crossref","is-referenced-by-count":3,"title":["LeGrad: An Explainability Method for Vision Transformers via Feature Formation Sensitivity"],"prefix":"10.1109","author":[{"given":"Walid","family":"Bousselham","sequence":"first","affiliation":[{"name":"University of Tuebingen & Goethe University,Frankfurt"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Angie","family":"Boggust","sequence":"additional","affiliation":[{"name":"MIT CSAIL"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sofian","family":"Chaybouti","sequence":"additional","affiliation":[{"name":"University of Tuebingen & Goethe University,Frankfurt"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hendrik","family":"Strobelt","sequence":"additional","affiliation":[{"name":"MIT-IBM Watson AI Lab"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hilde","family":"Kuehne","sequence":"additional","affiliation":[{"name":"University of Tuebingen & Goethe University,Frankfurt"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.acl-main.385"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0130140"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00198"},{"key":"ref4","article-title":"From colouring-in to pointillism: revisiting semantic segmentation supervision","volume-title":"arXiv preprint","author":"Benenson","year":"2022"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1145\/3491102.3501965"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00367"},{"key":"ref7","article-title":"Gifford. What made you do this? Understanding black-box decisions with sufficient input subsets.","volume-title":"Proceedings of the International Conference on Artificial Intelligence and Statistics (AISTATS). PMLR","author":"Carter","year":"2019"},{"key":"ref8","article-title":"Overinterpretation reveals image classification model pathologies","volume-title":"NeurIPS","author":"Carter","year":"2021"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00084"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00045"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00276"},{"key":"ref12","volume-title":"Vision transformers need registers. ICLR","author":"Darcet","year":"2024"},{"key":"ref13","article-title":"An image is worth $16 \\times 16$ words: Transformers for image recognition at scale","volume-title":"ICLR","author":"Dosovitskiy","year":"2020"},{"key":"ref14","article-title":"Visualizing higher-layer features of a deep network","volume-title":"Technical Report, Univerist\u00e9 de Montr\u00e9al","author":"Erhan","year":"2009"},{"key":"ref15","article-title":"Interpreting clip\u2019s image representation via text-based decomposition","volume-title":"ICLR","author":"Gandelsman","year":"2023"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2022.3218275"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01457"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.23915\/distill.00030"},{"key":"ref19","article-title":"Contrastive audio-visual masked autoencoder","volume-title":"ICLR","author":"Gong","year":"2023"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01246"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01231-1_40"},{"key":"ref22","article-title":"Natural language descriptions of deep visual features","volume-title":"ICLR","author":"Hernandez","year":"2021"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00368"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00505"},{"key":"ref25","article-title":"Set transformer: A framework for attention-based permutation-invariant neural networks","volume-title":"ICML.","author":"Lee","year":"2019"},{"key":"ref26","article-title":"A unified approach to interpreting model predictions","volume-title":"NeurIPS","author":"Lundberg","year":"2017"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.58248\/pn633"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1145\/3583558"},{"key":"ref29","article-title":"RISE: Randomized input sampling for explanation of black-box models","volume-title":"BMVC.","author":"Petsiuk","year":"2018"},{"key":"ref30","article-title":"Learning transferable visual models from natural language supervision","volume-title":"ICML.","author":"Radford","year":"2021"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/N16-3020"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-015-0816-y"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.74"},{"key":"ref34","article-title":"Deep inside convolutional networks: Visualising image classification models and saliency maps","volume-title":"ICLR, (Workshop Track)","author":"Simonyan","year":"2014"},{"key":"ref35","article-title":"Smoothgrad: removing noise by adding noise","volume-title":"arXiv preprint","author":"Smilkov","year":"2017"},{"key":"ref36","article-title":"Riedmiller","volume-title":"Striving for simplicity: The all convolutional net.","author":"Springenberg","year":"2015"},{"key":"ref37","article-title":"Axiomatic attribution for deep networks","volume-title":"ICML. JMLR. org","author":"Sundararajan","year":"2017"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/p19-1580"},{"key":"ref39","article-title":"Demystifying clip data","volume-title":"ICLR","author":"Xu","year":"2023"},{"key":"ref40","article-title":"Coca: Contrastive captioners are image-text foundation models","volume-title":"arXiv preprint","author":"Yu","year":"2022"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-10590-1_53"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01100"}],"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\/11444655.pdf?arnumber=11444655","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T04:53:37Z","timestamp":1777611217000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11444655\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":42,"URL":"https:\/\/doi.org\/10.1109\/iccv51701.2025.01891","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}