Picture for Xiaoming Xu

Xiaoming Xu

LLIA -- Enabling Low-Latency Interactive Avatars: Real-Time Audio-Driven Portrait Video Generation with Diffusion Models

Add code
Jun 06, 2025
Viaarxiv icon

SageAttention2++: A More Efficient Implementation of SageAttention2

Add code
May 28, 2025
Viaarxiv icon

SageAttention3: Microscaling FP4 Attention for Inference and An Exploration of 8-Bit Training

Add code
May 16, 2025
Viaarxiv icon

BEM: Balanced and Entropy-based Mix for Long-Tailed Semi-Supervised Learning

Add code
Apr 01, 2024
Viaarxiv icon

EfficientRep:An Efficient Repvgg-style ConvNets with Hardware-aware Neural Network Design

Add code
Feb 01, 2023
Viaarxiv icon

YOLOv6 v3.0: A Full-Scale Reloading

Add code
Jan 13, 2023
Viaarxiv icon

Semi-MAE: Masked Autoencoders for Semi-supervised Vision Transformers

Add code
Jan 04, 2023
Viaarxiv icon

YOLOv6: A Single-Stage Object Detection Framework for Industrial Applications

Add code
Sep 07, 2022
Figure 1 for YOLOv6: A Single-Stage Object Detection Framework for Industrial Applications
Figure 2 for YOLOv6: A Single-Stage Object Detection Framework for Industrial Applications
Figure 3 for YOLOv6: A Single-Stage Object Detection Framework for Industrial Applications
Figure 4 for YOLOv6: A Single-Stage Object Detection Framework for Industrial Applications
Viaarxiv icon