Michele Magno

Quantized Visual Geometry Grounded Transformer

Post-Training Quantization for Video Matting

QuantSparse: Comprehensively Compressing Video Diffusion Transformer with Model Quantization and Attention Sparsification

$\text{S}^2$Q-VDiT: Accurate Quantized Video Diffusion Transformer with Salient Data and Sparse Token Distillation

BinaryDM: Accurate Weight Binarization for Efficient Diffusion Models

SliM-LLM: Salience-Driven Mixed-Precision Quantization for Large Language Models

SliM-LLM: Salience-Driven Mixed-Precision Quantization for Large Language Models

Q-VDiT: Towards Accurate Quantization and Distillation of Video-Generation Diffusion Transformers