collaborators

11 papers

cs.LG2026

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism

Zhichen Zeng, Chi-Chih Chang, Jiayi Wang +10

Mixture-of-experts (MoE) architectures enable trillion-parameter LLMs with sparsely activated experts. Expert parallelism (EP) is a widely adopted MoE training strategy, but it suf…

cs.LG2026

SRT: Accelerating Reinforcement Learning via Speculative Rollout with Tree-Structured Cache

Chi-Chih Chang, Siqi Zhu, Zhichen Zeng +5

We present Speculative Rollout with Tree-Structured Cache (SRT), a simple, model-free approach to accelerate on-policy reinforcement learning (RL) for language models without sacri…

cs.DC2025

Mesh-Attention: A New Communication-Efficient Distributed Attention with Improved Data Locality

Sirui Chen, Jingji Chen, Siqi Zhu +3

Distributed attention is essential for scaling large language models (LLMs) to long contexts, yet existing methods either have limited parallelism or incur high communication costs…

cs.LG2025

MegaScale-MoE: Large-Scale Communication-Efficient Training of Mixture-of-Experts Models in Production

Chao Jin, Ziheng Jiang, Zhihao Bai +16

We present MegaScale-MoE, a production system tailored for the efficient training of large-scale mixture-of-experts (MoE) models. MoE emerges as a promising architecture to scale l…

cs.DC2025

MegaScale-Infer: Serving Mixture-of-Experts at Scale with Disaggregated Expert Parallelism

Ruidong Zhu, Ziheng Jiang, Chao Jin +17

Mixture-of-Experts (MoE) showcases tremendous potential to scale large language models (LLMs) with enhanced performance and reduced computational complexity. However, its sparsely…

cs.DC2025

SwiftSpec: Ultra-Low Latency LLM Decoding by Scaling Asynchronous Speculative Decoding

Ziyi Zhang, Ziheng Jiang, Chengquan Jiang +5

Low-latency decoding for large language models (LLMs) is crucial for applications like chatbots and code assistants, yet generating long outputs remains slow in single-query settin…