activity
20242026
collaborators

6 papers

cs.CV2026

Efficient Difficulty-Aware Dynamic Routing for Diffusion-Based Real-World Image Super-Resolution

Xue Wu, Kang Zhao, Kafeng Wang +4

Diffusion-based methods have achieved impressive performance in real-world image super-resolution (Real-ISR) by leveraging large pre-trained stable diffusion (SD) models as powerfu…

cs.CV2026

PreciseCache: Precise Feature Caching for Efficient and High-fidelity Video Generation

Jiangshan Wang, Kang Zhao, Jiayi Guo +5

High computational costs and slow inference hinder the practical application of video generation models. While prior works accelerate the generation process through feature caching…

cs.LG2025

Maximum Redundancy Pruning: A Principle-Driven Layerwise Sparsity Allocation for LLMs

Chang Gao, Kang Zhao, Runqi Wang +2

Large language models (LLMs) have demonstrated impressive capabilities, but their enormous size poses significant challenges for deployment in real-world applications. To address t…

cs.LG2024

Accelerating Transformer Pre-training with 2:4 Sparsity

Yuezhou Hu, Kang Zhao, Weiyu Huang +2

Training large transformers is slow, but recent innovations on GPU architecture give us an advantage. NVIDIA Ampere GPUs can execute a fine-grained 2:4 sparse matrix multiplication…

cs.LG2024

1-Bit FQT: Pushing the Limit of Fully Quantized Training to 1-bit

Chang Gao, Jianfei Chen, Kang Zhao +2

Fully quantized training (FQT) accelerates the training of deep neural networks by quantizing the activations, weights, and gradients into lower precision. To explore the ultimate…

cs.LG2024

Jetfire: Efficient and Accurate Transformer Pretraining with INT8 Data Flow and Per-Block Quantization

Haocheng Xi, Yuxiang Chen, Kang Zhao +3

Pretraining transformers are generally time-consuming. Fully quantized training (FQT) is a promising approach to speed up pretraining. However, most FQT methods adopt a quantize-co…