{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,9]],"date-time":"2026-06-09T16:16:46Z","timestamp":1781021806337,"version":"3.54.1"},"reference-count":25,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,10,26]],"date-time":"2025-10-26T00:00:00Z","timestamp":1761436800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,10,26]],"date-time":"2025-10-26T00:00:00Z","timestamp":1761436800000},"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":[],"published-print":{"date-parts":[[2025,10,26]]},"DOI":"10.1109\/iccad66269.2025.11240633","type":"proceedings-article","created":{"date-parts":[[2025,11,20]],"date-time":"2025-11-20T18:39:34Z","timestamp":1763663974000},"page":"1-8","source":"Crossref","is-referenced-by-count":1,"title":["COBRA: Algorithm-Architecture Co-optimized Binary Transformer Accelerator for Edge Inference"],"prefix":"10.1109","author":[{"given":"Ye","family":"Qiao","sequence":"first","affiliation":[{"name":"University of California,Department of Electrical Engineering and Computer Science,Irvine,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhiheng","family":"Chen","sequence":"additional","affiliation":[{"name":"University of California,Department of Electrical Engineering and Computer Science,Irvine,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yian","family":"Wang","sequence":"additional","affiliation":[{"name":"University of California,Department of Electrical Engineering and Computer Science,Irvine,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yifan","family":"Zhang","sequence":"additional","affiliation":[{"name":"University of California,Department of Electrical Engineering and Computer Science,Irvine,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yunzhe","family":"Deng","sequence":"additional","affiliation":[{"name":"University of California,Department of Electrical Engineering and Computer Science,Irvine,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sitao","family":"Huang","sequence":"additional","affiliation":[{"name":"University of California,Department of Electrical Engineering and Computer Science,Irvine,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","first-page":"2","article-title":"BERT: Pre-training of deep bidirectional transformers for language understanding","volume-title":"Proceedings of naacL-HLT","volume":"1","author":"Kenton"},{"key":"ref2","article-title":"An image is worth 16x16 words: Transformers for image recognition at scale","author":"Alexey","year":"2020"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.20472\/EFC.2023.018.004"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA48891.2023.10161317"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/EMC2-NIPS53020.2019.00016"},{"key":"ref6","first-page":"24 101","article-title":"A fast post-training pruning framework for transformers","volume":"35","author":"Kwon","year":"2022","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1145\/3020078.3021741"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1145\/3431920.3439296"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/ICIT48603.2022.10002796"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/TETC.2023.3237778"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.acl-long.334"},{"key":"ref12","article-title":"BiBERT: Accurate fully binarized BERT","author":"Qin","year":"2022"},{"key":"ref13","first-page":"14 303","article-title":"BiT: Robustly binarized multi-distilled transformer","volume":"35","author":"Liu","year":"2022","journal-title":"Advances in neural information processing systems"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.emnlp-main.37"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2017.2682138"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/CoolChips.2018.8373076"},{"key":"ref17","article-title":"Vaqf: Fully automatic software-hardware co-design framework for low-bit vision transformer","author":"Sun","year":"2022"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/ISCAS58744.2024.10558636"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1145\/3676536.3676733"},{"key":"ref20","article-title":"Flashattention: fast and memory-efficient exact attention with io-awareness","volume-title":"Proceedings of the 36th International Conference on Neural Information Processing Systems","author":"Dao"},{"key":"ref21","article-title":"GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding","author":"Wang","year":"2019"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.23919\/DATE51398.2021.9474043"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/DAC56929.2023.10247678"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/ASP-DAC58780.2024.10473976"},{"key":"ref25","article-title":"Hamming weight \u2014 wikipedia, the free encyclopedia"}],"event":{"name":"2025 IEEE\/ACM International Conference On Computer Aided Design (ICCAD)","location":"Munich, Germany","start":{"date-parts":[[2025,10,26]]},"end":{"date-parts":[[2025,10,30]]}},"container-title":["2025 IEEE\/ACM International Conference On Computer Aided Design (ICCAD)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11240608\/11240621\/11240633.pdf?arnumber=11240633","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,21]],"date-time":"2025-11-21T05:42:12Z","timestamp":1763703732000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11240633\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,26]]},"references-count":25,"URL":"https:\/\/doi.org\/10.1109\/iccad66269.2025.11240633","relation":{},"subject":[],"published":{"date-parts":[[2025,10,26]]}}}