{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T20:17:56Z","timestamp":1783455476265,"version":"3.55.0"},"reference-count":85,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"8","license":[{"start":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:00:00Z","timestamp":1785542400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:00:00Z","timestamp":1785542400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:00:00Z","timestamp":1785542400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100000923","name":"Australian Research Council","doi-asserted-by":"publisher","award":["DP230101753"],"award-info":[{"award-number":["DP230101753"]}],"id":[{"id":"10.13039\/501100000923","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100000923","name":"Australian Research Council","doi-asserted-by":"publisher","award":["DE200101610"],"award-info":[{"award-number":["DE200101610"]}],"id":[{"id":"10.13039\/501100000923","id-type":"DOI","asserted-by":"publisher"}]},{"name":"CSIRO’s Science Leader Project","award":["R-91559"],"award-info":[{"award-number":["R-91559"]}]},{"name":"Responsible AI Research Centre"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Pattern Anal. Mach. Intell."],"published-print":{"date-parts":[[2026,8]]},"DOI":"10.1109\/tpami.2026.3677028","type":"journal-article","created":{"date-parts":[[2026,3,24]],"date-time":"2026-03-24T19:48:04Z","timestamp":1774381684000},"page":"9078-9092","source":"Crossref","is-referenced-by-count":0,"title":["Understanding the Effects of Projectors in Knowledge Distillation"],"prefix":"10.1109","volume":"48","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-5672-5313","authenticated-orcid":false,"given":"Yudong","family":"Chen","sequence":"first","affiliation":[{"name":"School of Electrical Engineering and Computer Science, The University of Queensland, St Lucia, QLD, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5414-8276","authenticated-orcid":false,"given":"Sen","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Electrical Engineering and Computer Science, The University of Queensland, St Lucia, QLD, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8160-1796","authenticated-orcid":false,"given":"Jiajun","family":"Liu","sequence":"additional","affiliation":[{"name":"School of Electrical Engineering and Computer Science, The University of Queensland, St Lucia, QLD, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xuwei","family":"Xu","sequence":"additional","affiliation":[{"name":"School of Electrical Engineering and Computer Science, The University of Queensland, St Lucia, QLD, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Frank","family":"de Hoog","sequence":"additional","affiliation":[{"name":"CSIRO Data61, Brisbane, QLD, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Brano","family":"Kusy","sequence":"additional","affiliation":[{"name":"CSIRO Data61, Brisbane, QLD, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zi","family":"Huang","sequence":"additional","affiliation":[{"name":"School of Electrical Engineering and Computer Science, The University of Queensland, St Lucia, QLD, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","first-page":"3519","article-title":"Similarity of neural network representations revisited","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Kornblith","year":"2019"},{"key":"ref2","first-page":"1321","article-title":"On calibration of modern neural networks","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Guo","year":"2017"},{"key":"ref3","first-page":"1097","article-title":"Imagenet classification with deep convolutional neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Krizhevsky","year":"2012"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.4324\/9781410605337-29"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01167"},{"key":"ref7","article-title":"MobileNets: Efficient convolutional neural networks for mobile vision applications","author":"Howard","year":"2017"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00716"},{"key":"ref9","article-title":"Distilling the knowledge in a neural network","author":"Hinton","year":"2015"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2021.3055564"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2022.3200344"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2020.3001940"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00201"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.52202\/068431-0878"},{"key":"ref15","first-page":"10347","article-title":"Training data-efficient image transformers & distillation through attention","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Touvron","year":"2021"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.52202\/068431-0666"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-021-01453-z"},{"key":"ref18","article-title":"Contrastive representation distillation","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Tian","year":"2020"},{"key":"ref19","first-page":"22158","article-title":"Comprehensive knowledge distillation with causal intervention","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Deng","year":"2021"},{"key":"ref20","article-title":"Knowledge distillation via softmax regression representation learning","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Yang","year":"2021"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-024-02192-7"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58529-7_39"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-20083-0_4"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00497"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i9.16969"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i8.16865"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00489"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01165"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00454"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i04.5963"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2021.3067100"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2022.3170728"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2022.3212905"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00396"},{"key":"ref35","article-title":"MEAL V2: Boosting vanilla ResNet-50 to 80% top-1 accuracy on ImageNet without tricks","author":"Shen","year":"2020"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-20053-3_39"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01065"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.52202\/068431-2443"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.52202\/068431-0044"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00145"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00511"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.754"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2020.2970494"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58610-2_2"},{"key":"ref45","article-title":"Feature kernel distillation","volume-title":"Proc. Int. Conf. Learn. Representations","author":"He","year":"2021"},{"key":"ref46","article-title":"Distilling representational similarity using centered kernel alignment (CKA)","volume-title":"Proc. Brit. Mach. Vis. Conf.","author":"Saha","year":"2022"},{"key":"ref47","article-title":"FitNets: Hints for thin deep nets","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Romero","year":"2015"},{"key":"ref48","first-page":"2765","article-title":"Paraphrasing complex network: Network compression via factor transfer","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Kim","year":"2018"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v38i5.28219"},{"key":"ref50","first-page":"15720","article-title":"VkD: Improving knowledge distillation using orthogonal projections","volume-title":"Proc. IEEE\/CVF Conf. Comput. Vis. Pattern Recognit.","author":"Miles","year":"2024"},{"key":"ref51","article-title":"Paying more attention to attention: Improving the performance of convolutional neural networks via attention transfer","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Komodakis","year":"2017"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01549"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.5555\/3495724.3497510"},{"key":"ref54","article-title":"Better teacher better student: Dynamic prior knowledge for knowledge distillation","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Qiu","year":"2023"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58545-7_34"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1038\/nature21056"},{"key":"ref57","article-title":"End to end learning for self-driving cars","author":"Bojarski","year":"2016"},{"key":"ref58","first-page":"15682","article-title":"Revisiting the calibration of modern neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Minderer","year":"2021"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2023.3270288"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.634"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00409"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-015-0816-y"},{"key":"ref63","article-title":"Learning multiple layers of features from tiny images","author":"Krizhevsky","year":"2009","journal-title":"Tech. Rep."},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.243"},{"key":"ref65","article-title":"Very deep convolutional networks for large-scale image recognition","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Simonyan","year":"2015"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00474"},{"key":"ref67","article-title":"Gaussian error linear units (GELUs)","author":"Hendrycks","year":"2016"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.24818\/ida-ql\/2019.5"},{"key":"ref69","article-title":"An image is worth 16 \u00d7 16 words: Transformers for image recognition at scale","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Dosovitskiy","year":"2021"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.123"},{"key":"ref71","article-title":"Exact solutions to the nonlinear dynamics of learning in deep linear neural networks","author":"Saxe","year":"2013"},{"key":"ref72","volume-title":"Causal Inference in Statistics: A Primer","author":"Glymour","year":"2016"},{"key":"ref73","first-page":"655","article-title":"Causal intervention for weakly-supervised semantic segmentation","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Zhang","year":"2020"},{"key":"ref74","first-page":"2734","article-title":"Interventional few-shot learning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Yue","year":"2020"},{"key":"ref75","article-title":"Representation learning with contrastive predictive coding","author":"Oord","year":"2018"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58621-8_45"},{"key":"ref77","article-title":"The caltech-ucsd birds-200-2011 dataset","author":"Wah","year":"2011"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.1109\/ICCVW.2013.77"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01352"},{"key":"ref80","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.324"},{"key":"ref81","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.106"},{"key":"ref82","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-10602-1_48"},{"key":"ref83","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.350"},{"key":"ref84","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-014-0733-5"},{"key":"ref85","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1802.02611"}],"container-title":["IEEE Transactions on Pattern Analysis and Machine Intelligence"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/34\/11595778\/11455346.pdf?arnumber=11455346","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T19:44:28Z","timestamp":1783453468000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11455346\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8]]},"references-count":85,"journal-issue":{"issue":"8"},"URL":"https:\/\/doi.org\/10.1109\/tpami.2026.3677028","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":[[2026,8]]}}}