IEEE

EdgeDiff: Energy-Efficient Multi-Modal Few-Step Diffusion Model Accelerator Using Mixed-Precision and Reordered Group Quantization

Abstract: Recent advances in diffusion models (DMs)—such as few-step denoising and multi-modal conditioning—have significantly improved computational efficiency and functional flexibility, but they ...
IEEE

Theoretical Model, Experimental Survey, and Optimization of Pulse Parameters for 4H-SiC Drift Step Recovery Diode

Abstract: The drift step recovery diode (DSRD) is one of the most popular superfast pulse power switches. The pulse parameter model of DSRD is important to improve the characteristics of the pulsed ...
GitHub

SimVQ: Addressing Representation Collapse in Vector Quantized Models with One Linear Layer

imagenet └── train/ ├── n01440764 ├── n01440764_10026.JPEG ├── n01440764_10027.JPEG ├── ... ├── n01443537 ...
GitHub

Distilled Decoding 1: One-step Sampling of Image Auto-regressive Models with Flow Matching

This is the official repository of paper Distilled Decoding 1: One-step Sampling of Image Auto-regressive Models with Flow Matching. We propose Distilled Decoding (DD) to distill a pre-trained image ...