Donglin Yang -

Not registered as user

Name:Donglin Yang

Affiliation:Nvidia Corporation

Contributions

2026

Principles and Practice of Parallel Programming

Author of JanusQuant: Accurate and Efficient 2-bit KV Cache Quantization for Long-Context Inference within the Main Conference-track

2025

Principles and Practice of Parallel Programming

Author of Harnessing Inter-GPU Shared Memory for Seamless MoE Communication-Computation Fusion within the Main Conference-track