TrainingActive
TWIST
High-Efficiency LLM Training System via Strand Interleaving on NVIDIA Hopper GPUs.
llmtrainingdistributed-systemshopper
MLSys @ USTC
Projects
Systems for distributed LLM training and efficient AI infrastructure.
SystemsActive
Adacluster
Adaptive clustering system for large-scale ML workloads with dynamic resource scheduling.
ml-systemsschedulingresource-management