Wenxiang Lin, Xinglin Pan, Lin Zhang 0059, Shaohuai Shi, Xuan Wang 0002, Xiaowen Chu 0001. HierMoE: Accelerating MoE Training with Hierarchical Token Deduplication and Expert Swap. In IEEE INFOCOM 2026 - IEEE Conference on Computer Communications, Tokyo, Japan, May 18-21, 2026. pages 1-10, IEEE, 2026. [doi]
Abstract is missing.