{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,15]],"date-time":"2026-01-15T22:48:17Z","timestamp":1768517297876,"version":"3.49.0"},"reference-count":21,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,12,14]],"date-time":"2025-12-14T00:00:00Z","timestamp":1765670400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,12,14]],"date-time":"2025-12-14T00:00:00Z","timestamp":1765670400000},"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":[[2025,12,14]]},"DOI":"10.1109\/icpads67057.2025.11323171","type":"proceedings-article","created":{"date-parts":[[2026,1,14]],"date-time":"2026-01-14T20:36:54Z","timestamp":1768423014000},"page":"1-8","source":"Crossref","is-referenced-by-count":0,"title":["Fine-Grained State Sharding and Communication Pipelining for Partially Sharded Data Parallelism"],"prefix":"10.1109","author":[{"given":"Weimin","family":"Li","sequence":"first","affiliation":[{"name":"SKLP, Institute of Computing Technology, CAS,Beijing,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yuzhong","family":"Sun","sequence":"additional","affiliation":[{"name":"SKLP, Institute of Computing Technology, CAS,Beijing,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Lin","family":"Meng","sequence":"additional","affiliation":[{"name":"SKLP, Institute of Computing Technology, CAS,Beijing,China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.14778\/3415478.3415530"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/SC41405.2020.00024"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM.2019.8737367"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1145\/3394486.3406703"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.14778\/3611540.3611569"},{"key":"ref6","first-page":"551","article-title":"{Zero-offload}: Democratizing {billion-scale} model training","volume-title":"2021 USENIX Annual Technical Conference (USENIX ATC 21)","author":"Ren","year":"2021"},{"key":"ref7","article-title":"Zero++: Extremely efficient collective communication for large model training","volume-title":"The Twelfth International Conference on Learning Representations","author":"Wang","year":"2024"},{"key":"ref8","first-page":"3252","article-title":"Error feedback fixes signsgd and other gradient compression schemes","volume-title":"International Conference on Machine Learning.","author":"Karimireddy","year":"2019"},{"key":"ref9","first-page":"181","article-title":"Poseidon: An efficient communication architecture for distributed deep learning on {GPU} clusters","volume-title":"2017 USENIX Annual Technical Conference (USENIX ATC 17)","author":"Zhang","year":"2017"},{"key":"ref10","author":"Yang","year":"2024","journal-title":"Qwen2. 5 technical report"},{"key":"ref11","volume-title":"Qwen2.5 model page on modelscope"},{"key":"ref12","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3458817.3476205","article-title":"Zeroinfinity: Breaking the gpu memory wall for extreme scale deep learning","volume-title":"Proceedings of the international conference for high performance computing, networking, storage and analysis","author":"Rajbhandari","year":"2021"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1145\/3341301.3359642"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM42981.2021.9488803"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.5555\/2685048.2685095"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1145\/505202.505215"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2022.3219819"},{"key":"ref18","author":"Levy","year":"2016","journal-title":"The power of normalization: Faster evasion of saddle points"},{"key":"ref19","author":"Feng","year":"2024","journal-title":"Echo: Simulating distributed training at scale"},{"key":"ref20","article-title":"Parameter-efficient fine-tuning for large models: A comprehensive survey","volume":"2024","author":"Han","year":"2024","journal-title":"Transactions on Machine Learning Research"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1145\/1375527.1375580"}],"event":{"name":"2025 IEEE 31th International Conference on Parallel and Distributed Systems (ICPADS)","location":"Hefei, China","start":{"date-parts":[[2025,12,14]]},"end":{"date-parts":[[2025,12,18]]}},"container-title":["2025 IEEE 31th International Conference on Parallel and Distributed Systems (ICPADS)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11322805\/11322871\/11323171.pdf?arnumber=11323171","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,15]],"date-time":"2026-01-15T07:49:16Z","timestamp":1768463356000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11323171\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,14]]},"references-count":21,"URL":"https:\/\/doi.org\/10.1109\/icpads67057.2025.11323171","relation":{},"subject":[],"published":{"date-parts":[[2025,12,14]]}}}