Wang Zhe Mark

BLOB-Q: Boosting Low Bit ViT Quantization via Global Optimization on Model Distortion

Quantization with Purpose: Loss-Aware Bit Allocation for Gradient Compression

CERSA: Cumulative Energy-Retaining Subspace Adaptation for Memory-Efficient Fine-Tuning

Distribution-Aware Diffusion Model Quantization via Distortion Minimization

DM3D: Parameter-Efficient and Lossless 3D Object Detection via Distortion Minimization