conftrace_

expert parallelism

9 papers

Explore in graph

Also known as

EP

Co-occurring keywords · papers mentioning both

mixture of expert (7) large language model (2) pipeline parallelism (2) quality degradation (1) token routing (1) modality-aware capacity (1) prefill-decode disaggregation (1) all-to-all communication (1) buffer fusion (1) cpu gpu hybrid (1)

Papers

Achieving Cloud-Grade SLOs for Local Mixture-of-Experts Inference through CPU–GPU Hybrid Design OSDI 2026 MACS: Modality-Aware Capacity Scaling for Efficient Multimodal MoE Inference ACL 2026 RCMoE: A Communication-Efficient Random Compression Framework for Resource-Constrained Mixture-of-Experts Training AAAI 2026 SMIDT: High-Performance Inference Framework for MoE Models with Dynamic Top-K Routing AAAI 2026 SwiftEP: Accelerating MoE Inference with Buffer Fusion and TMA Offloading NSDI 2026 UEP: Portable Expert-Parallel Communication OSDI 2026 DICE: Staleness-Centric Optimizations for Parallel Diffusion MoE Inference ICCV 2025 Occult: Optimizing Collaborative Communications across Experts for Accelerated Parallel MoE Training and Inference ICML 2025 Shortcut-connected Expert Parallelism for Accelerating Mixture of Experts ICML 2025