{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,18]],"date-time":"2026-03-18T13:23:17Z","timestamp":1773840197536,"version":"3.50.1"},"reference-count":38,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"9","license":[{"start":{"date-parts":[[2024,9,1]],"date-time":"2024-09-01T00:00:00Z","timestamp":1725148800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2024,9,1]],"date-time":"2024-09-01T00:00:00Z","timestamp":1725148800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2024,9,1]],"date-time":"2024-09-01T00:00:00Z","timestamp":1725148800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100014188","name":"MSIT (Ministry of Science and ICT), Korea","doi-asserted-by":"publisher","award":["IITP-2020-0-01847"],"award-info":[{"award-number":["IITP-2020-0-01847"]}],"id":[{"id":"10.13039\/501100014188","id-type":"DOI","asserted-by":"publisher"}]},{"name":"IITP"},{"name":"BK21 FOUR"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Comput."],"published-print":{"date-parts":[[2024,9]]},"DOI":"10.1109\/tc.2024.3404051","type":"journal-article","created":{"date-parts":[[2024,5,21]],"date-time":"2024-05-21T17:23:45Z","timestamp":1716312225000},"page":"2248-2261","source":"Crossref","is-referenced-by-count":2,"title":["ToEx: Accelerating Generation Stage of Transformer-Based Language Models via Token-Adaptive Early Exit"],"prefix":"10.1109","volume":"73","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-3557-8526","authenticated-orcid":false,"given":"Myeonggu","family":"Kang","sequence":"first","affiliation":[{"name":"School of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, South Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-7722-8702","authenticated-orcid":false,"given":"Junyoung","family":"Park","sequence":"additional","affiliation":[{"name":"School of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, South Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0382-4032","authenticated-orcid":false,"given":"Hyein","family":"Shin","sequence":"additional","affiliation":[{"name":"School of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, South Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5943-1599","authenticated-orcid":false,"given":"Jaekang","family":"Shin","sequence":"additional","affiliation":[{"name":"School of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, South Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9585-4591","authenticated-orcid":false,"given":"Lee-Sup","family":"Kim","sequence":"additional","affiliation":[{"name":"School of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, South Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"key":"ref2","article-title":"Exploring the limits of transfer learning with a unified text-to-text transformer","author":"Raffel","year":"2023"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.acl-main.703"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/tnnls.2020.2979670"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00254"},{"key":"ref6","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1016\/j.aiopen.2020.11.001","article-title":"Neural machine translation: A review of methods, resources, and tools","volume":"1","author":"Tan","year":"2020","journal-title":"AI Open"},{"key":"ref7","article-title":"Teaching machines to read and comprehend","author":"Hermann","year":"2015"},{"key":"ref8","article-title":"BERT loses patience: Fast and robust inference with early exit","author":"Zhou","year":"2020"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.acl-main.204"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1145\/3466752.3480095"},{"key":"ref11","article-title":"BERT: Pre-training of deep bidirectional transformers for language understanding","author":"Devlin","year":"2018"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.emnlp-demos.6"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.472"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46493-0_32"},{"key":"ref15","article-title":"A3: Accelerating attention mechanisms in neural networks with approximation","author":"Ham","year":"2020"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/hpca51647.2021.00018"},{"key":"ref17","first-page":"131","article-title":"Findings of the 2016 conference on machine translation","volume-title":"Proc. 1st Conf. Mach. Transl.","author":"Boja","year":"2016"},{"key":"ref18","doi-asserted-by":"crossref","first-page":"1073","DOI":"10.18653\/v1\/P17-1099","article-title":"Get to the point: Summarization with pointer-generator networks","volume-title":"Proc. 55th Annu. Meeting Assoc. Comput. Linguistics (Volume 1: Long Papers)","author":"See","year":"2017"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D18-1206"},{"key":"ref20","first-page":"8024","article-title":"PyTorch: An imperative style, high-performance deep learning library","volume-title":"Proc. Adv. Neural Inf. Process. Syst. 32 (NeurIPS)","author":"Paszke","year":"2019"},{"key":"ref21","first-page":"311","article-title":"Bleu: A method for automatic evaluation of machine translation","volume-title":"Proc. 40th Annu. Meeting Assoc. Comput. Linguistics","author":"Papineni","year":"2002"},{"key":"ref22","first-page":"74","article-title":"ROUGE: A package for automatic evaluation of summaries","volume-title":"Text Summarization Branches Out","author":"Lin","year":"2004"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/LCA.2020.2973991"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/3085572"},{"key":"ref25","article-title":"Confident adaptive language modeling","author":"Schuster","year":"2022"},{"key":"ref26","first-page":"91","article-title":"BERxiT: Early exiting for BERT with better fine-tuning and extension to regression","volume-title":"Proc. 16th Conf. Eur. Chapter Assoc. Comput. Linguistics: Main Volume. Online: Assoc. Comput. Linguistics,","author":"Xin","year":"2021"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/icpr.2016.7900006"},{"key":"ref28","article-title":"Adaptive neural networks for fast test-time prediction","author":"Bolukbasi","year":"2017"},{"key":"ref29","first-page":"14128","article-title":"Early exit with disentangled representation and equiangular tight frame","volume-title":"Proc. Findings Assoc. Comput. Linguistics (ACL),","author":"Ji","year":"2023"},{"key":"ref30","article-title":"LUT-GEMM: Quantized matrix multiplication based on LUTs for efficient inference in large-scale generative language models","author":"Park","year":"2024"},{"key":"ref31","article-title":"AWQ: Activation-aware weight quantization for LLM compression and acceleration","author":"Lin","year":"2024"},{"key":"ref32","article-title":"Memory-efficient fine-tuning of compressed large language models via sub-4-bit integer quantization","author":"Kim","year":"2023"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/micro50266.2020.00071"},{"key":"ref34","article-title":"SqueezeLLM: Dense-and-sparse quantization","author":"Kim","year":"2024"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.sustainlp-1.6"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.emnlp-main.829"},{"key":"ref37","first-page":"521","article-title":"Orca: A distributed serving system for transformer-based generative models","volume-title":"Proc. 16th USENIX Symp. Operating Syst. Des. Implementation (OSDI 22)","author":"Yu","year":"2022"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1145\/3600006.3613165"}],"container-title":["IEEE Transactions on Computers"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/12\/10631806\/10535998.pdf?arnumber=10535998","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,8,11]],"date-time":"2024-08-11T04:13:06Z","timestamp":1723349586000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10535998\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,9]]},"references-count":38,"journal-issue":{"issue":"9"},"URL":"https:\/\/doi.org\/10.1109\/tc.2024.3404051","relation":{},"ISSN":["0018-9340","1557-9956","2326-3814"],"issn-type":[{"value":"0018-9340","type":"print"},{"value":"1557-9956","type":"electronic"},{"value":"2326-3814","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,9]]}}}