June 25, 2026 HybridEP: Scaling Expert Parallelism to Cross-Datacenter Scenario Hybrid Expert/Data Transmission for MoE Training moe expert-parallelism distributed-training cross-datacenter communication-optimization compression
June 22, 2026 Federation of Experts: Communication Efficient Distributed Inference for Large Language Models 通过将 MoE 架构重构为多个独立的专家集群，消除 all-to-all 通信瓶颈，实现高效的分布式推理 llm mixture-of-experts distributed-inference communication-efficiency expert-parallelism