{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,14]],"date-time":"2026-04-14T15:24:56Z","timestamp":1776180296575,"version":"3.50.1"},"reference-count":53,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"4","license":[{"start":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T00:00:00Z","timestamp":1775001600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T00:00:00Z","timestamp":1775001600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T00:00:00Z","timestamp":1775001600000},"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","award":["62322117"],"award-info":[{"award-number":["62322117"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62371365"],"award-info":[{"award-number":["62371365"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["U22B2014"],"award-info":[{"award-number":["U22B2014"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["U24B20136"],"award-info":[{"award-number":["U24B20136"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Circuits Syst. Video Technol."],"published-print":{"date-parts":[[2026,4]]},"DOI":"10.1109\/tcsvt.2025.3636885","type":"journal-article","created":{"date-parts":[[2025,11,25]],"date-time":"2025-11-25T18:30:48Z","timestamp":1764095448000},"page":"5246-5257","source":"Crossref","is-referenced-by-count":0,"title":["MultiGS: Multi-Dimensional Information-Aware Gradient Sparsification"],"prefix":"10.1109","volume":"36","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-7782-8184","authenticated-orcid":false,"given":"Jitao","family":"Ma","sequence":"first","affiliation":[{"name":"State Key Laboratory of Integrated Services Networks, Xidian University, Xi’an, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Donglai","family":"Liu","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Integrated Services Networks, Xidian University, Xi’an, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8310-024X","authenticated-orcid":false,"given":"Weiying","family":"Xie","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Integrated Services Networks, Xidian University, Xi’an, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Haowei","family":"Li","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Integrated Services Networks, Xidian University, Xi’an, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0234-6270","authenticated-orcid":false,"given":"Yunsong","family":"Li","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Integrated Services Networks, Xidian University, Xi’an, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2351-4461","authenticated-orcid":false,"given":"Leyuan","family":"Fang","sequence":"additional","affiliation":[{"name":"College of Electrical and Information Engineering, Hunan University, Changsha, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2024.3494820"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2025.3570707"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2025.3560615"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2024.3375511"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2023.3281983"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2025.3550803"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/SC41405.2020.00024"},{"key":"ref8","article-title":"Megatron-LM: Training multi-billion parameter language models using model parallelism","author":"Shoeybi","year":"2019","journal-title":"arXiv:1909.08053"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2024.3407131"},{"key":"ref10","first-page":"4447","article-title":"Sparsified SGD with memory","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"31","author":"Stich"},{"key":"ref11","article-title":"Variance reduction with sparse gradients","author":"Elibol","year":"2020","journal-title":"arXiv:2001.09623"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i04.5793"},{"key":"ref13","article-title":"Don\u2019t use large mini-batches, use local SGD","author":"Lin","year":"2018","journal-title":"arXiv:1808.07217"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/JSAC.2022.3192050"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D17-1045"},{"key":"ref16","article-title":"Deep gradient compression: Reducing the communication bandwidth for distributed training","author":"Lin","year":"2017","journal-title":"arXiv:1712.01887"},{"key":"ref17","first-page":"12593","article-title":"CSER: Communication-efficient SGD with error reset","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Xie"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1016\/S0893-6080(98)00116-6"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref20","article-title":"Very deep convolutional networks for large-scale image recognition","author":"Simonyan","year":"2014","journal-title":"arXiv:1409.1556"},{"key":"ref21","article-title":"Using the output embedding to improve language models","volume-title":"arXiv:1608.05859","author":"Press","year":"2016"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00474"},{"key":"ref23","article-title":"Better plain ViT baselines for ImageNet-1k","author":"Beyer","year":"2022","journal-title":"arXiv:2205.01580"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2014-274"},{"key":"ref25","first-page":"560","article-title":"signSGD: Compressed optimisation for non-convex problems","volume-title":"Proc. 35th Int. Conf. Mach. Learn.","volume":"80","author":"Bernstein"},{"key":"ref26","first-page":"3252","article-title":"Error feedback fixes SignSGD and other gradient compression schemes","volume-title":"Proc. Int. Conf. Mach. Learn.","volume":"97","author":"Karimireddy"},{"key":"ref27","first-page":"1707","article-title":"QSGD: Communication-efficient SGD via gradient quantization and encoding","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Alistarh"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00552"},{"key":"ref29","first-page":"9872","article-title":"ATOMO: Communication-efficient learning via atomic sparsification","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Wang"},{"key":"ref30","first-page":"14269","article-title":"PowerSGD: Practical low-rank gradient compression for distributed optimization","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Vogels"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/ICDCS.2019.00220"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/JSAIT.2020.2985917"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2021.3084806"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01189"},{"key":"ref35","first-page":"11450","article-title":"Communication-efficient distributed blockwise momentum SGD with error-feedback","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Zheng"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v32i1.11728"},{"key":"ref37","first-page":"18102","article-title":"Errorcompensatedx: Error compensation for variance reduced algorithms","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"34","author":"Tang"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/5.726791"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1145\/3065386"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.4324\/9781410605337-29"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2020.3010627"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/34.192463"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2002.804424"},{"key":"ref44","article-title":"Learning multiple layers of features from tiny images","author":"Krizhevsky","year":"2009"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.21236\/ADA273556"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/W18-5446"},{"key":"ref48","article-title":"LLaMA: Open and efficient foundation language models","author":"Touvron","year":"2023","journal-title":"arXiv:2302.13971"},{"key":"ref49","article-title":"OPT: Open pre-trained transformer language models","author":"Zhang","year":"2022","journal-title":"arXiv:2205.01068"},{"key":"ref50","article-title":"Gemma 2: Improving open language models at a practical size","author":"Team","year":"2024","journal-title":"arXiv:2408.00118"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D13-1170"},{"key":"ref52","article-title":"Neural network acceptability judgments","author":"Warstadt","year":"2018","journal-title":"arXiv:1805.12471"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.187"}],"container-title":["IEEE Transactions on Circuits and Systems for Video Technology"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/76\/11475579\/11267482.pdf?arnumber=11267482","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,7]],"date-time":"2026-04-07T20:03:35Z","timestamp":1775592215000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11267482\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4]]},"references-count":53,"journal-issue":{"issue":"4"},"URL":"https:\/\/doi.org\/10.1109\/tcsvt.2025.3636885","relation":{},"ISSN":["1051-8215","1558-2205"],"issn-type":[{"value":"1051-8215","type":"print"},{"value":"1558-2205","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4]]}}}