arxiv:2505.19151
Cheng Shenggan
shenggan
AI & ML interests
Machine Learning System, High Performance Computing
Recent Activity
authored
a paper
11 days ago
Hanayo: Harnessing Wave-like Pipeline Parallelism for Enhanced Large
Model Training Efficiency
authored
a paper
11 days ago
HeteGen: Heterogeneous Parallel Inference for Large Language Models on
Resource-Constrained Devices
authored
a paper
11 days ago
DSP: Dynamic Sequence Parallelism for Multi-Dimensional Transformers
Organizations
None yet
models
0
None public yet
datasets
0
None public yet