{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,12]],"date-time":"2026-05-12T23:26:28Z","timestamp":1778628388401,"version":"3.51.4"},"reference-count":28,"publisher":"IEEE","license":[{"start":{"date-parts":[[2021,6,6]],"date-time":"2021-06-06T00:00:00Z","timestamp":1622937600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2021,6,6]],"date-time":"2021-06-06T00:00:00Z","timestamp":1622937600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2021,6,6]],"date-time":"2021-06-06T00:00:00Z","timestamp":1622937600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100012226","name":"Fundamental Research Funds for the Central Universities","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100012226","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2021,6,6]]},"DOI":"10.1109\/icassp39728.2021.9414572","type":"proceedings-article","created":{"date-parts":[[2021,5,13]],"date-time":"2021-05-13T19:53:45Z","timestamp":1620935625000},"page":"7713-7717","source":"Crossref","is-referenced-by-count":16,"title":["Elbert: Fast Albert with Confidence-Window Based Early Exit"],"prefix":"10.1109","author":[{"given":"Keli","family":"Xie","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Siyuan","family":"Lu","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Meiqi","family":"Wang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhongfeng","family":"Wang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref10","article-title":"Once-for-all: Train one network and specialize it for efficient deployment","author":"cai","year":"2019","journal-title":"ICLRE"},{"key":"ref11","doi-asserted-by":"crossref","DOI":"10.18653\/v1\/2020.repl4nlp-1.18","article-title":"Compressing bert: Studying the effects of weight pruning on transfer learning","author":"gordon","year":"2020"},{"key":"ref12","first-page":"14014","article-title":"Are sixteen heads really better than one?","author":"michel","year":"2019","journal-title":"Advances in neural information processing systems"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i05.6409"},{"key":"ref14","doi-asserted-by":"crossref","DOI":"10.1109\/EMC2-NIPS53020.2019.00016","article-title":"Q8bert: Quantized 8bit bert","author":"zafrir","year":"2019"},{"key":"ref15","article-title":"Distilling task-specific knowledge from bert into simple neural networks","author":"tang","year":"2019"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D19-1441"},{"key":"ref17","article-title":"Distilbert, a distilled version of bert: smaller, faster, cheaper and lighter","author":"sanh","year":"2019"},{"key":"ref18","article-title":"Tinybert: Distilling bert for natural language understanding","author":"jiao","year":"2019"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/ICPR.2016.7900006"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P19-3007"},{"key":"ref4","article-title":"Deep compression: Compressing deep neural networks with pruning, trained quantization and huffman coding","author":"han","year":"2015"},{"key":"ref27","first-page":"142","article-title":"Learning word vectors for sentiment analysis","author":"maas","year":"2011","journal-title":"ACL"},{"key":"ref3","article-title":"Xlnet: Generalized autoregressive pretraining for language understanding","author":"yang","year":"2019","journal-title":"Advances in neural information processing systems"},{"key":"ref6","article-title":"Distilling the knowledge in a neural network","author":"hinton","year":"2015"},{"key":"ref5","article-title":"Compressing deep convolutional networks using vector quantization","author":"gong","year":"2014"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D19-1445"},{"key":"ref7","article-title":"Albert: A lite bert for self-supervised learning of language representations","author":"lan","year":"2019","journal-title":"ICLRE"},{"key":"ref2","article-title":"Roberta: A robustly optimized bert pretraining approach","author":"liu","year":"2019"},{"key":"ref9","first-page":"3301","article-title":"Shallow-deep networks: Understanding and mitigating network over-thinking","author":"kaya","year":"2019","journal-title":"ICML"},{"key":"ref1","article-title":"Bert: Pre-training of deep bidirectional transformers for language understanding","author":"devlin","year":"2019","journal-title":"NAACL"},{"key":"ref20","article-title":"Reducing transformer depth on demand with structured dropout","author":"fan","year":"2019","journal-title":"ICLRE"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.acl-main.593"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.acl-main.204"},{"key":"ref24","first-page":"178","article-title":"Dynexit: A dynamic early-exit strategy for deep residual net-works","author":"wang","year":"2019","journal-title":"SIPS"},{"key":"ref23","doi-asserted-by":"crossref","DOI":"10.18653\/v1\/2020.acl-main.537","article-title":"Fastbert: a self-distilling bert with adaptive inference time","author":"liu","year":"2020"},{"key":"ref26","article-title":"Character-level convolutional networks for text classification","author":"zhang","year":"2015","journal-title":"Advances in neural information processing systems"},{"key":"ref25","article-title":"Glue: A multi-task benchmark and analysis platform for natural language understanding","author":"wang","year":"2019","journal-title":"ICLRE"}],"event":{"name":"ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","location":"Toronto, ON, Canada","start":{"date-parts":[[2021,6,6]]},"end":{"date-parts":[[2021,6,11]]}},"container-title":["ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9413349\/9413350\/09414572.pdf?arnumber=9414572","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,12,27]],"date-time":"2022-12-27T08:29:53Z","timestamp":1672129793000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9414572\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,6,6]]},"references-count":28,"URL":"https:\/\/doi.org\/10.1109\/icassp39728.2021.9414572","relation":{},"subject":[],"published":{"date-parts":[[2021,6,6]]}}}