{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,25]],"date-time":"2025-11-25T14:09:33Z","timestamp":1764079773068,"version":"3.37.3"},"reference-count":25,"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:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2022,1,1]],"date-time":"2022-01-01T00:00:00Z","timestamp":1640995200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2022,1,1]],"date-time":"2022-01-01T00:00:00Z","timestamp":1640995200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"National Key R&D Program of China","award":["2018YFB1005103"],"award-info":[{"award-number":["2018YFB1005103"]}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62006138","61925601","61772302"],"award-info":[{"award-number":["62006138","61925601","61772302"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Huawei Noah’s Ark Lab"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE\/ACM Trans. Audio Speech Lang. Process."],"published-print":{"date-parts":[[2022]]},"DOI":"10.1109\/taslp.2022.3153257","type":"journal-article","created":{"date-parts":[[2022,3,7]],"date-time":"2022-03-07T20:49:53Z","timestamp":1646686193000},"page":"958-967","source":"Crossref","is-referenced-by-count":7,"title":["Dynamic Multi-Branch Layers for On-Device Neural Machine Translation"],"prefix":"10.1109","volume":"30","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2426-6220","authenticated-orcid":false,"given":"Zhixing","family":"Tan","sequence":"first","affiliation":[{"name":"Department of Computer Science and Technology, Tsinghua University, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zeyuan","family":"Yang","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Technology, Tsinghua University, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Meng","family":"Zhang","sequence":"additional","affiliation":[{"name":"Huawei Noah’s Ark Lab, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7000-1792","authenticated-orcid":false,"given":"Qun","family":"Liu","sequence":"additional","affiliation":[{"name":"Huawei Noah’s Ark Lab, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6011-6115","authenticated-orcid":false,"given":"Maosong","family":"Sun","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Technology, Tsinghua University, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3087-242X","authenticated-orcid":false,"given":"Yang","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Technology, Tsinghua University, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1146\/annurev.neuro.26.041002.131047"},{"article-title":"Controlling computation versus quality for neural sequence models","year":"2020","author":"Bapna","key":"ref2"},{"article-title":"Estimating or propagating gradients through stochastic neurons for conditional computation","year":"2013","author":"Bengio","key":"ref3"},{"key":"ref4","article-title":"Conditional computation in neural networks for faster models","volume-title":"Proc. Conf. Learn. Representations, Workshop Track.","author":"Bengio","year":"2016"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-39593-2_1"},{"article-title":"Exponentially increasing the capacity-to-computation ratio for conditional computation in deep learning","year":"2014","author":"Cho","key":"ref6"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.findings-emnlp.433"},{"key":"ref8","article-title":"Low-rank approximations for conditional feedforward computation in deep neural networks","volume-title":"Proc. Conf. Learn. Representations, Workshop Track.","author":"Davis","year":"2014"},{"key":"ref9","article-title":"Learning factored representations in a deep mixture of experts","volume-title":"Proc. Conf. Learn. Representations, Workshop Track.","author":"Eigen","year":"2014"},{"article-title":"Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity","year":"2021","author":"Fedus","key":"ref10"},{"key":"ref11","article-title":"The lottery ticket hypothesis: Finding sparse, trainable neural networks","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Frankle","year":"2019"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D16-1139"},{"key":"ref13","article-title":"Adam: A method for stochastic optimization","volume-title":"Proc. Int. Conf. Learn. Representations.","author":"Kingma","year":"2015"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/SOCC49529.2020.9524802"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.3115\/1073083.1073135"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/W18-6319"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.findings-emnlp.1"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P16-1162"},{"key":"ref19","article-title":"Outrageously large neural networks: The sparsely-gated mixture-of-experts layer","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Shazeer","year":"2017"},{"key":"ref20","first-page":"5877","article-title":"The evolved transformer","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"So","year":"2019"},{"key":"ref21","first-page":"496","article-title":"Sequence to sequence learning with neural networks","volume-title":"Proc. Neural Inf. Process. Syst.","author":"Sutskever","year":"2014"},{"key":"ref22","first-page":"5998","article-title":"Attention is all you need","volume-title":"Proc. Neural Inf. Process. Syst.","author":"Vaswani","year":"2017"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.acl-main.686"},{"key":"ref24","article-title":"Lite transformer with long-short range attention","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Wu","year":"2020"},{"key":"ref25","article-title":"To prune, or not to prune: Exploring the efficacy of pruning for model compression","volume-title":"Proc. Conf. Learn. Representations, Workshop Track.","author":"Zhu","year":"2018"}],"container-title":["IEEE\/ACM Transactions on Audio, Speech, and Language Processing"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/6570655\/9657755\/09729651.pdf?arnumber=9729651","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,1,17]],"date-time":"2024-01-17T23:03:04Z","timestamp":1705532584000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9729651\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022]]},"references-count":25,"URL":"https:\/\/doi.org\/10.1109\/taslp.2022.3153257","relation":{},"ISSN":["2329-9290","2329-9304"],"issn-type":[{"type":"print","value":"2329-9290"},{"type":"electronic","value":"2329-9304"}],"subject":[],"published":{"date-parts":[[2022]]}}}