Jidong Zhai

Papers on Lattice

Total citations

Topics

h-index

Research focus

Architecture Design (Transformers, SSMs, MoE) (2)Distributed Systems & Hardware (2)Training Efficiency & Optimization (1)Inference & Quantization (1)

Frequent co-authors

Size Zheng (1)Xuegui Zheng (1)Li-Wen Chang (1)Chia-chi Hsieh (1)Chia-chi Hsieh (1)Zan Zong (1)Zan Zong (1)Xinyang Chen (1)

Papers (2)

Apr 21, 2026

Size Zheng +3Apr 21, 2026

UniEP: Unified Expert-Parallel MoE MegaKernel for LLM Training

Training MoE models just got a whole lot faster and more reliable: UniEP's MegaKernels and deterministic token ordering unlock significant speedups without sacrificing numerical stability.

Size Zheng, Xuegui Zheng, Li-Wen Chang +1

Architecture Design (Transformers, SSMs, MoE)Distributed Systems & Hardware Training Efficiency & Optimization

Feb 18, 2026

Tsinghua AIFeb 18, 2026·also USTB

FlowPrefill: Decoupling Preemption from Prefill Scheduling Granularity to Mitigate Head-of-Line Blocking in LLM Serving

LLM serving can achieve 5.6x higher throughput without sacrificing latency by decoupling preemption granularity from scheduling frequency.

Chia-chi Hsieh, Chia-chi Hsieh, Zan Zong +8

Architecture Design (Transformers, SSMs, MoE)Distributed Systems & Hardware Inference & Quantization

Jidong Zhai

Research focus

Frequent co-authors

Papers (2)

Search

Jidong Zhai

Research focus

Frequent co-authors

Papers (2)