{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,13]],"date-time":"2026-08-13T06:29:25Z","timestamp":1786602565591,"version":"build-2736575974"},"reference-count":72,"publisher":"IEEE","license":[{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"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":[[2026,5]]},"DOI":"10.1109\/icde65706.2026.00107","type":"proceedings-article","created":{"date-parts":[[2026,8,12]],"date-time":"2026-08-12T19:23:42Z","timestamp":1786562622000},"page":"1377-1390","source":"Crossref","is-referenced-by-count":0,"title":["SSFusion: Tensor Fusion with Selective Sparsification for Efficient Distributed DNN Training"],"prefix":"10.1109","author":[{"given":"Zhangqiang","family":"Ming","sequence":"first","affiliation":[{"name":"Huazhong University of Science and Technology","place":["Wuhan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rui","family":"Wang","sequence":"additional","affiliation":[{"name":"Huazhong University of Science and Technology","place":["Wuhan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuchong","family":"Hu","sequence":"additional","affiliation":[{"name":"Huazhong University of Science and Technology","place":["Wuhan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuanhao","family":"Shu","sequence":"additional","affiliation":[{"name":"Innovation Research Institute of Cethik Group Co., Ltd.","place":["Hangzhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wenxiang","family":"Zhou","sequence":"additional","affiliation":[{"name":"Huazhong University of Science and Technology","place":["Wuhan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xinjue","family":"Zheng","sequence":"additional","affiliation":[{"name":"Huazhong University of Science and Technology","place":["Wuhan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dan","family":"Feng","sequence":"additional","affiliation":[{"name":"Huazhong University of Science and Technology","place":["Wuhan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"key":"ref2","article-title":"Bert: Pre-training of deep bidirectional transformers for language understanding","author":"Devlin","year":"2018","journal-title":"arXiv preprint"},{"key":"ref3","first-page":"1877","article-title":"Language models are few-shot learners","author":"Brown","year":"2020","journal-title":"Advances in Neural Information Processing Systems"},{"issue":"8","key":"ref4","first-page":"9","article-title":"Language models are unsupervised multitask learners","volume":"1","author":"Radford","year":"2019","journal-title":"OpenAI blog"},{"key":"ref5","article-title":"Horovod: fast and easy distributed deep learning in tensorflow","author":"Sergeev","year":"2018","journal-title":"arXiv preprint"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1145\/3477132.3483553"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.14778\/3415478.3415530"},{"key":"ref8","first-page":"463","article-title":"A unified architecture for accelerating distributed dnn training in heterogeneous gpu\/cpu clusters","volume-title":"Proceedings of the 14th USENIX Conference on Operating Systems Design and Implementation","author":"Jiang","year":"2020"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1145\/3394486.3406703"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/ICDE51399.2021.00040"},{"key":"ref11","article-title":"Bringing hpc techniques to deep learning","author":"Gibiansky","year":"2017","journal-title":"Baidu Research, Tech. Rep."},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1145\/3552326.3567505"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM.2019.8737367"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1145\/3485447.3511981"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM53939.2023.10228871"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM53939.2023.10228922"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM41043.2020.9155269"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/ICDCS57875.2023.00054"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1145\/3625549.3658678"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1145\/3503221.3508399"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2019\/473"},{"key":"ref22","article-title":"Understanding top-k sparsification in distributed deep learning","author":"Shi","year":"2019","journal-title":"arXiv preprint"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1016\/j.jpdc.2019.05.016"},{"key":"ref24","first-page":"297","article-title":"An efficient statistical-based gradient compression technique for distributed training systems","volume-title":"Proceedings of Machine Learning and Systems","author":"Abdelmoniem","year":"2021"},{"key":"ref25","first-page":"8133","article-title":"Rethinking gradient sparsification as total error minimization","author":"Sahu","year":"2021","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref26","volume-title":"Byteps","year":"2025"},{"key":"ref27","volume-title":"Deepspeed","year":"2025"},{"key":"ref28","article-title":"Cupcake: A compression scheduler for scalable communication-efficient distributed training","volume-title":"Proceedings of the Sixth Conference on Machine Learning and Systems (MLSys\u2019 23). Proceedings of the Sixth Conference on Machine Learning and Systems (MLSys\u2019 23)","author":"Wang","year":"2023"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1145\/3731545.3731581"},{"key":"ref30","article-title":"Mxnet: A flexible and efficient machine learning library for heterogeneous distributed systems","author":"Chen","year":"2015","journal-title":"arXiv preprint"},{"key":"ref31","first-page":"1027","article-title":"Accelerating collective communication in data parallel training across deep learning frameworks","volume-title":"19th USENIX Symposium on Networked Systems Design and Implementation (NSDI 22)","author":"Romero","year":"2022"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM42981.2021.9488803"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/ICDE60146.2024.00142"},{"key":"ref34","volume-title":"Horovod","year":"2025"},{"key":"ref35","volume-title":"Nvidia nccl","year":"2025"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1145\/3018743.3018769"},{"key":"ref37","first-page":"1","article-title":"Poseidon: An efficient communication architecture for distributed deep learning on gpu clusters","volume-title":"USENIX Annual Technical Conference","volume":"1","author":"Zhang","year":"2017"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2021.3052862"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1145\/3503222.3507778"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1145\/3341301.3359642"},{"key":"ref41","first-page":"785","article-title":"Scaling distributed machine learning with in-network aggregation","volume-title":"18th USENIX Symposium on Networked Systems Design and Implementation (NSDI 21)","author":"Sapio","year":"2021"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/ICDCS51616.2021.00060"},{"key":"ref43","first-page":"623","article-title":"dpro: A generic performance diagnosis and optimization toolkit for expediting distributed dnn training","volume-title":"Proceedings of Machine Learning and Systems","volume":"4","author":"Hu","year":"2022"},{"key":"ref44","article-title":"Deep gradient compression: Reducing the communication bandwidth for distributed training","volume-title":"International Conference on Learning Representations","author":"Lin","year":"2018"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1145\/3452296.3472904"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1145\/3295500.3356222"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/ICDCS.2019.00220"},{"key":"ref48","first-page":"21150","article-title":"Deepreduce: A sparse-tensor communication framework for federated deep learning","author":"Xu","year":"2021","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/ICCD58817.2023.00096"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2022.3154387"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2022.3230938"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/ICDE55515.2023.00183"},{"key":"ref53","first-page":"1306","article-title":"Gradient sparsification for communication-efficient distributed optimization","author":"Wangni","year":"2018","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref54","first-page":"5977","article-title":"The convergence of sparsified gradient methods","author":"Alistarh","year":"2018","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/MASS52906.2021.00025"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.5555\/3294771.3294915"},{"key":"ref57","first-page":"5325","article-title":"Error compensated quantized sgd and its applications to large-scale distributed optimization","volume-title":"International Conference on Machine Learning","author":"Wu","year":"2018"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref59","volume-title":"Learning multiple layers of features from tiny images","author":"Krizhevsky","year":"2009"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P18-2124"},{"key":"ref61","article-title":"Regularizing and optimizing 1stm language models","author":"Merity","year":"2017","journal-title":"arXiv preprint"},{"key":"ref62","first-page":"249","article-title":"Understanding the difficulty of training deep feedforward neural networks","volume-title":"Proceedings of the thirteenth international conference on artificial intelligence and statistics. JMLR Workshop and Conference Proceedings","author":"Glorot","year":"2010"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i04.5793"},{"key":"ref64","article-title":"An image is worth 16\u00d716 words: Transformers for image recognition at scale","author":"Dosovitskiy","year":"2020","journal-title":"arXiv preprint"},{"key":"ref65","first-page":"4452","article-title":"Sparsified sgd with memory","author":"Stich","year":"2018","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref66","first-page":"652","article-title":"On the utility of gradient compression in distributed training systems","volume-title":"Proceedings of Machine Learning and Systems","volume":"4","author":"Agarwal","year":"2022"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-540-24685-5_1"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1145\/3673038.3673140"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref70","article-title":"Very deep convolutional networks for large-scale image recognition","author":"Simonyan","year":"2014","journal-title":"arXiv preprint"},{"key":"ref71","article-title":"Pointer sentinel mixture models","author":"Merity","year":"2016","journal-title":"arXiv preprint"},{"key":"ref72","volume-title":"Bloom: A 176b-parameter open-access multilingual language model","author":"Le Scao","year":"2023"}],"event":{"name":"2026 IEEE 42nd International Conference on Data Engineering (ICDE)","location":"Montreal, QC, Canada","start":{"date-parts":[[2026,5,4]]},"end":{"date-parts":[[2026,5,8]]}},"container-title":["2026 IEEE 42nd International Conference on Data Engineering (ICDE)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11629178\/11629165\/11629542.pdf?arnumber=11629542","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,13]],"date-time":"2026-08-13T06:04:19Z","timestamp":1786601059000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11629542\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5]]},"references-count":72,"URL":"https:\/\/doi.org\/10.1109\/icde65706.2026.00107","relation":{},"subject":[],"published":{"date-parts":[[2026,5]]}}}