{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,20]],"date-time":"2026-10-20T16:51:03Z","timestamp":1792515063380,"version":"4.5.0"},"reference-count":76,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2022,1,1]],"date-time":"2022-01-01T00:00:00Z","timestamp":1640995200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"funder":[{"DOI":"10.13039\/501100008530","name":"European Regional Development Fund","doi-asserted-by":"publisher","award":["KK.01.1.1.01.0009 (DATACROSS)"],"award-info":[{"award-number":["KK.01.1.1.01.0009 (DATACROSS)"]}],"id":[{"id":"10.13039\/501100008530","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Access"],"published-print":{"date-parts":[[2022]]},"DOI":"10.1109\/access.2022.3169772","type":"journal-article","created":{"date-parts":[[2022,4,22]],"date-time":"2022-04-22T15:34:23Z","timestamp":1650641663000},"page":"47011-47030","source":"Crossref","is-referenced-by-count":32,"title":["Toward Practical Usage of the Attention Mechanism as a Tool for Interpretability"],"prefix":"10.1109","volume":"10","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-5227-5397","authenticated-orcid":false,"given":"Martin","family":"Tutek","sequence":"first","affiliation":[{"name":"Faculty of Electrical Engineering and Computing, University of Zagreb, Zagreb, Croatia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8942-5301","authenticated-orcid":false,"given":"Jan","family":"Snajder","sequence":"additional","affiliation":[{"name":"Faculty of Electrical Engineering and Computing, University of Zagreb, Zagreb, Croatia"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","first-page":"1097","article-title":"ImageNet classification with deep convolutional neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst. (NIPS)","volume":"25","author":"Krizhevsky"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref3","first-page":"5998","article-title":"Attention is all you need","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Vaswani"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/W18-5450"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-021-03819-2"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.8.1735"},{"key":"ref7","article-title":"Neural machine translation by jointly learning to align and translate","author":"Bahdanau","year":"2014","journal-title":"arXiv:1409.0473"},{"key":"ref8","article-title":"Attention-over-attention neural networks for reading comprehension","author":"Cui","year":"2016","journal-title":"arXiv:1607.04423"},{"key":"ref9","article-title":"Bidirectional attention flow for machine comprehension","author":"Seo","year":"2016","journal-title":"arXiv:1611.01603"},{"key":"ref10","article-title":"A structured self-attentive sentence embedding","author":"Lin","year":"2017","journal-title":"arXiv:1703.03130"},{"key":"ref11","article-title":"Rethinking attention with performers","author":"Choromanski","year":"2020","journal-title":"arXiv:2009.14794"},{"key":"ref12","article-title":"Language models are few-shot learners","author":"Brown","year":"2020","journal-title":"arXiv:2005.14165"},{"key":"ref13","article-title":"Structured attention networks","author":"Kim","year":"2017","journal-title":"arXiv:1702.00887"},{"key":"ref14","article-title":"Graph attention networks","author":"Veli\u010dkovi\u0107","year":"2017","journal-title":"arXiv:1710.10903"},{"key":"ref15","first-page":"10183","article-title":"Synthesizer: Rethinking self-attention for transformer models","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Tay"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P16-2096"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1145\/3351095.3375671"},{"key":"ref18","article-title":"Deep inside convolutional networks: Visualising image classification models and saliency maps","author":"Simonyan","year":"2013","journal-title":"arXiv:1312.6034"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/N16-1082"},{"key":"ref20","first-page":"3145","article-title":"Learning important features through propagating activation differences","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Shrikumar"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/W18-5454"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P18-1032"},{"key":"ref23","first-page":"3543","article-title":"Attention is not explanation","volume-title":"Proc. Conf. North American ACL, Hum. Lang. Technol.","volume":"1","author":"Jain"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/W19-4813"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.emnlp-main.574"},{"key":"ref26","article-title":"When can models learn from explanations? A formal framework for understanding the roles of explanation data","author":"Hase","year":"2021","journal-title":"arXiv:2102.02201"},{"key":"ref27","article-title":"Order in the court: Explainable AI methods prone to disagreement","author":"Neely","year":"2021","journal-title":"arXiv:2105.03287"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1145\/3411764.3445088"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1145\/3290605.3300234"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1145\/3313831.3376718"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-57959-7"},{"key":"ref32","article-title":"Byte pair encoding: A text compression scheme that accelerates pattern matching","author":"Shibata","year":"1999"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P16-1162"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v30i1.10362"},{"key":"ref35","first-page":"3111","article-title":"Distributed representations of words and phrases and their compositionality","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Mikolov"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.3115\/v1\/D14-1162"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.5555\/3045118.3045336"},{"key":"ref38","first-page":"3504","article-title":"Retain: An interpretable predictive model for healthcare using reverse time attention mechanism","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Choi"},{"key":"ref39","first-page":"1614","article-title":"From softmax to sparsemax: A sparse model of attention and multi-label classification","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Martins"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P17-1088"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/W18-5427"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D19-1002"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.repl4nlp-1.17"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.3115\/v1\/D14-1179"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1207\/s15516709cog1402_1"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.acl-main.386"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/n18-2017"},{"key":"ref48","article-title":"Learning to deceive with attention-based explanations","author":"Pruthi","year":"2019","journal-title":"arXiv:1909. 07913"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-019-0048-x"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1145\/3233231"},{"key":"ref51","first-page":"4768","article-title":"A unified approach to interpreting model predictions","volume-title":"Proc. 31st Int. Conf. neural Inf. Process. Syst.","author":"Lundberg"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.acl-main.385"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D16-1011"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P19-1284"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.acl-main.409"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P19-1282"},{"key":"ref57","article-title":"Attention interpretability across NLP tasks","author":"Vashishth","year":"2019","journal-title":"arXiv:1909.11218"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298594"},{"key":"ref59","article-title":"Highway networks","author":"Srivastava","year":"2015","journal-title":"arXiv:1505.00387"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.2307\/1267352"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1111\/j.2517-6161.1996.tb02080.x"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1111\/j.1467-9868.2005.00503.x"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1810.04805"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D19-1448"},{"key":"ref65","article-title":"Tying word vectors and word classifiers: A loss framework for language modeling","author":"Inan","year":"2016","journal-title":"arXiv:1611. 01462"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1108\/eb026584"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.1175\/1520-0493(1950)078<0001:VOFEIT>2.0.CO;2"},{"key":"ref68","first-page":"368","article-title":"The information bottleneck method","volume-title":"Proc. 37th Annu. Allerton Conf. Commun., Control Comput.","author":"Tishby"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1162\/089976603321780272"},{"key":"ref70","first-page":"5228","article-title":"Assessing generative models via precision and recall","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Sajjadi"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.1145\/1772690.1772862"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.1145\/1014052.1014073"},{"key":"ref73","first-page":"1631","article-title":"Recursive deep models for semantic compositionality over a sentiment treebank","volume-title":"Proc. 2013 Conf. Empirical Methods Natural Lang. Process.","author":"Socher"},{"key":"ref74","first-page":"142","article-title":"Learning word vectors for sentiment analysis","volume-title":"Proc. 49th Annu. Meet. Assoc. Comput. Linguistics, Hum. Lang. Technol.","author":"Maas"},{"key":"ref75","doi-asserted-by":"publisher","DOI":"10.1038\/sdata.2016.35"},{"key":"ref76","first-page":"1310","article-title":"On the difficulty of training recurrent neural networks","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Pascanu"}],"container-title":["IEEE Access"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/6287639\/9668973\/09762294.pdf?arnumber=9762294","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,1,22]],"date-time":"2024-01-22T16:07:42Z","timestamp":1705939662000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9762294\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022]]},"references-count":76,"URL":"https:\/\/doi.org\/10.1109\/access.2022.3169772","relation":{},"ISSN":["2169-3536"],"issn-type":[{"value":"2169-3536","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022]]}}}