{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,25]],"date-time":"2026-09-25T04:04:49Z","timestamp":1790309089931,"version":"4.1.0"},"reference-count":96,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"7","license":[{"start":{"date-parts":[[2025,7,1]],"date-time":"2025-07-01T00:00:00Z","timestamp":1751328000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,7,1]],"date-time":"2025-07-01T00:00:00Z","timestamp":1751328000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,7,1]],"date-time":"2025-07-01T00:00:00Z","timestamp":1751328000000},"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. Neural Netw. Learning Syst."],"published-print":{"date-parts":[[2025,7]]},"DOI":"10.1109\/tnnls.2024.3501335","type":"journal-article","created":{"date-parts":[[2024,12,4]],"date-time":"2024-12-04T14:12:12Z","timestamp":1733321532000},"page":"11887-11901","source":"Crossref","is-referenced-by-count":16,"title":["SinKD: Sinkhorn Distance Minimization for Knowledge Distillation"],"prefix":"10.1109","volume":"36","author":[{"ORCID":"https:\/\/orcid.org\/0009-0008-4677-8834","authenticated-orcid":false,"given":"Xiao","family":"Cui","sequence":"first","affiliation":[{"name":"Department of Electrical Engineering and Information Science, University of Science and Technology of China, Hefei, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0996-3984","authenticated-orcid":false,"given":"Yulei","family":"Qin","sequence":"additional","affiliation":[{"name":"Tencent YouTu Laboratory, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuting","family":"Gao","sequence":"additional","affiliation":[{"name":"Tencent YouTu Laboratory, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Enwei","family":"Zhang","sequence":"additional","affiliation":[{"name":"Tencent YouTu Laboratory, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zihan","family":"Xu","sequence":"additional","affiliation":[{"name":"Tencent YouTu Laboratory, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tong","family":"Wu","sequence":"additional","affiliation":[{"name":"Tencent YouTu Laboratory, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7998-0731","authenticated-orcid":false,"given":"Ke","family":"Li","sequence":"additional","affiliation":[{"name":"Tencent YouTu Laboratory, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8132-9083","authenticated-orcid":false,"given":"Xing","family":"Sun","sequence":"additional","affiliation":[{"name":"Tencent YouTu Laboratory, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1690-9836","authenticated-orcid":false,"given":"Wengang","family":"Zhou","sequence":"additional","affiliation":[{"name":"Department of Electrical Engineering and Information Science, University of Science and Technology of China, Hefei, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2188-3028","authenticated-orcid":false,"given":"Houqiang","family":"Li","sequence":"additional","affiliation":[{"name":"Department of Electrical Engineering and Information Science, University of Science and Technology of China, Hefei, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","article-title":"BERT: Pre-training of deep bidirectional transformers for language understanding","author":"Devlin","year":"2018","journal-title":"arXiv:1810.04805"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1907.11692"},{"key":"ref3","article-title":"Multitask prompted training enables zero-shot task generalization","author":"Sanh","year":"2021","journal-title":"arXiv:2110.08207"},{"issue":"8","key":"ref4","first-page":"9","article-title":"Language models are unsupervised multitask learners","volume":"1","author":"Radford","year":"2019","journal-title":"OpenAI Blog"},{"key":"ref5","first-page":"1877","article-title":"Language models are few-shot learners","volume-title":"Proc. NeurIPS","author":"Brown"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1503.02531"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.acl-main.251"},{"key":"ref8","article-title":"MiniLLM: Knowledge distillation of large language models","author":"Gu","year":"2023","journal-title":"arXiv:2306.08543"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.acl-long.605"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00874"},{"key":"ref11","first-page":"11920","article-title":"Mosaicking to distill: Knowledge distillation from out-of-domain data","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Fang"},{"key":"ref12","first-page":"214","article-title":"Wasserstein generative adversarial networks","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Arjovsky"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D16-1139"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2021\/362"},{"key":"ref15","article-title":"Towards principled methods for training generative adversarial networks","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Arjovsky"},{"key":"ref16","article-title":"F-GAN: Training generative neural samplers using variational divergence minimization","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Nowozin"},{"key":"ref17","first-page":"10957","article-title":"Training deep energy-based models with f-divergence minimization","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Yu"},{"key":"ref18","first-page":"2292","article-title":"Sinkhorn distances: Lightspeed computation of optimal transport","volume-title":"Proc. Neural Inf. Process. Syst.","volume":"26","author":"Cuturi"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1137\/1118101"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/W18-5446"},{"key":"ref21","article-title":"SuperGLUE: A stickier benchmark for general-purpose language understanding systems","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"32","author":"Wang"},{"key":"ref22","article-title":"GPT-Neo: Large scale autoregressive language modeling with mesh-tensorflow","author":"Black"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2024.3501335"},{"key":"ref24","article-title":"DistilBERT, a distilled version of BERT: Smaller, faster, cheaper and lighter","author":"Sanh","year":"2019","journal-title":"arXiv:1910.01108"},{"key":"ref25","article-title":"Well-read students learn better: On the importance of pre-training compact models","author":"Turc","year":"2019","journal-title":"arXiv:1908.08962"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.findings-emnlp.372"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.acl-long.471"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2022.3152732"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2022.3186807"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2024.3372833"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.acl-long.485"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.acl-short.97"},{"key":"ref33","first-page":"69490","article-title":"KD-Zero: Evolving knowledge distiller for any teacher\u2013student pairs","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"36","author":"Li"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01597"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D19-1441"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00143"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00409"},{"key":"ref38","first-page":"13292","article-title":"Learning student-friendly teacher networks for knowledge distillation","volume-title":"Proc. Annu. Conf. Neural Inf. Process. Syst. (NeurIPS)","author":"Park"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2023.3257878"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-540-71050-9_6"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2021.3109942"},{"key":"ref42","first-page":"5769","article-title":"Improved training of Wasserstein GANs","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"30","author":"Gulrajani"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1561\/2200000073"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2023.107207"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1016\/j.isatra.2021.12.037"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v36i6.20645"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2022.3213641"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/tnnls.2023.3327357"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/tnnls.2023.3326867"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2022.3217189"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2023.3296642"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/tnnls.2023.3331304"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2020.3028022"},{"key":"ref54","article-title":"Learning with a Wasserstein loss","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Frogner"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2023.3262351"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-33383-5_12"},{"key":"ref57","article-title":"Joint distribution optimal transportation for domain adaptation","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Courty"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v36i10.21359"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2023.3267126"},{"key":"ref60","article-title":"Selective cross-task distillation","author":"Lu","year":"2022","journal-title":"arXiv:2204.11526"},{"key":"ref61","article-title":"KNOT: Knowledge distillation using optimal transport for solving NLP tasks","author":"Bhardwaj","year":"2021","journal-title":"arXiv:2110.02432"},{"key":"ref62","first-page":"1608","article-title":"Learning generative models with Sinkhorn divergences","volume-title":"Proc. Int. Conf. Artif. Intell. Statist.","author":"Genevay"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1016\/j.compchemeng.2022.108122"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00290"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D13-1170"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/N18-1101"},{"key":"ref67","first-page":"9","article-title":"Automatically constructing a corpus of sentential paraphrases","volume-title":"Proc. Int. Workshop Paraphrasing","author":"Dolan"},{"key":"ref68","first-page":"8","article-title":"The fifth PASCAL recognizing textual entailment challenge","volume":"7","author":"Bentivogli","year":"2009","journal-title":"IEEE Trans. Autom. Control"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D16-1264"},{"key":"ref70","article-title":"Quora question pairs","author":"Chen","year":"2018"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/S17-2001"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.emnlp-demos.6"},{"key":"ref73","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00454"},{"key":"ref74","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.emnlp-main.30"},{"key":"ref75","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i04.5963"},{"key":"ref76","volume-title":"Learning from deep model via exploring local targets","author":"Shi","year":"2020"},{"key":"ref77","first-page":"5776","article-title":"MiniLM: Deep self-attention distillation for task-agnostic compression of pre-trained transformers","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"33","author":"Wang"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.acl-main.195"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.acl-long.71"},{"issue":"2","key":"ref80","first-page":"107","article-title":"The commitmentbank: Investigating projection in naturally occurring discourse","volume-title":"Proc. Sinn Bedeutung","volume":"23","author":"De Marneffe"},{"key":"ref81","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.acl-demo.9"},{"key":"ref82","article-title":"Tailoring language generation models under total variation distance","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Ji"},{"key":"ref83","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.acl-long.502"},{"key":"ref84","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref85","article-title":"Contrastive representation distillation","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Tian"},{"key":"ref86","article-title":"FitNets: Hints for thin deep nets","author":"Romero","year":"2014","journal-title":"arXiv:1412.6550"},{"key":"ref87","article-title":"Paying more attention to attention: Improving the performance of convolutional neural networks via attention transfer","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Zagoruyko"},{"key":"ref88","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00145"},{"key":"ref89","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00511"},{"key":"ref90","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00938"},{"key":"ref91","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01252-6_17"},{"key":"ref92","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.33013779"},{"key":"ref93","article-title":"Paraphrasing complex network: Network compression via factor transfer","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"31","author":"Kim"},{"key":"ref94","article-title":"Very deep convolutional networks for large-scale image recognition","author":"Simonyan","year":"2014","journal-title":"arXiv:1409.1556"},{"key":"ref95","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00716"},{"key":"ref96","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01264-9_8"}],"container-title":["IEEE Transactions on Neural Networks and Learning Systems"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/5962385\/11073756\/10777837.pdf?arnumber=10777837","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,14]],"date-time":"2026-05-14T19:58:16Z","timestamp":1778788696000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10777837\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7]]},"references-count":96,"journal-issue":{"issue":"7"},"URL":"https:\/\/doi.org\/10.1109\/tnnls.2024.3501335","relation":{},"ISSN":["2162-237X","2162-2388"],"issn-type":[{"value":"2162-237X","type":"print"},{"value":"2162-2388","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7]]}}}