7 papers · 1 filter
HELLoRA: Hot Experts Layer-Level Low-Rank Adaptation for Mixture-of-Experts Models
Jia Wei, Zhonghao Zhang, Ping Chen +5
Low-Rank Adaptation (LoRA) dominates parameter-efficient fine-tuning of large language models, yet most variants target dense architectures. Mixture-of-Experts (MoE) models scale p…
SageAttention2++: A More Efficient Implementation of SageAttention2
Jintao Zhang, Xiaoming Xu, Jia Wei +5
The efficiency of attention is critical because its time complexity grows quadratically with sequence length. SageAttention2 addresses this by utilizing quantization to accelerate…
SageAttention3: Microscaling FP4 Attention for Inference and An Exploration of 8-Bit Training
Jintao Zhang, Jia Wei, Pengle Zhang +6
The efficiency of attention is important due to its quadratic time complexity. We enhance the efficiency of attention through two key contributions: First, we leverage the new FP4…
Accurate INT8 Training Through Dynamic Block-Level Fallback
Pengle Zhang, Jia Wei, Jintao Zhang +2
Transformer models have achieved remarkable success across various AI applications but face significant training costs. Low-bit training, such as INT8 training, can leverage comput…
SpargeAttention: Accurate and Training-free Sparse Attention Accelerating Any Model Inference
Jintao Zhang, Chendong Xiang, Haofeng Huang +4
An efficient attention implementation is essential for large models due to its quadratic time complexity. Fortunately, attention commonly exhibits sparsity, i.e., many values in th…
SageAttention2: Efficient Attention with Thorough Outlier Smoothing and Per-thread INT4 Quantization
Jintao Zhang, Haofeng Huang, Pengle Zhang +3
Although quantization for linear layers has been widely used, its application to accelerate the attention process remains limited. To further enhance the efficiency of attention co…