2 papers
cs.PF2026
Optimus: A Generic Operator-Level PyTorch Model Transformation Framework
Menglu Yu, Jiaqi Xu, Yuzhen Huang +19
In large-scale industrial applications, deep learning models that power recommendation and ranking have complex and diverse model architectures. These models are continuously devel…
cs.LG2024
Flex Attention: A Programming Model for Generating Optimized Attention Kernels
Juechu Dong, Boyuan Feng, Driss Guessous +2
Over the past 7 years, attention has become one of the most important primitives in deep learning. The primary approach to optimize attention is FlashAttention, which fuses the ope…