5 citations · 5 across the 7 of their papers we have counts for
7 papers
VideoSEMA: a scalable and efficient Mamba-like attention for video understanding
Nhat Thanh Tran, Fanghui Xue, Shuai Zhang +4
We present for video understanding (classification) a split space-time attention model, VideoSEMA, consisting of a scalable and efficient Mamba-like attention (SEMA) block in space…
SEMA: a Scalable and Efficient Mamba like Attention via Token Localization and Averaging
Nhat Thanh Tran, Fanghui Xue, Shuai Zhang +4
Attention is the critical component of a transformer. Yet the quadratic computational complexity of vanilla full attention in the input size and the inability of its linear attenti…
AFIDAF: Alternating Fourier and Image Domain Adaptive Filters as an Efficient Alternative to Attention in ViTs
Yunling Zheng, Zeyi Xu, Fanghui Xue +5
We propose and demonstrate an alternating Fourier and image domain filtering approach for feature extraction as an efficient alternative to build a vision backbone without using th…
A Proximal Algorithm for Network Slimming
Kevin Bui, Fanghui Xue, Fredrick Park +2
As a popular channel pruning method for convolutional neural networks (CNNs), network slimming (NS) has a three-stage process: (1) it trains a CNN with regularization appl…
Searching Intrinsic Dimensions of Vision Transformers
Fanghui Xue, Biao Yang, Yingyong Qi +1
It has been shown by many researchers that transformers perform as well as convolutional neural networks in many computer vision tasks. Meanwhile, the large computational costs of…
RARTS: An Efficient First-Order Relaxed Architecture Search Method
Fanghui Xue, Yingyong Qi, Jack Xin
Differentiable architecture search (DARTS) is an effective method for data-driven neural network design based on solving a bilevel optimization problem. Despite its success in many…
Learning Sparse Neural Networks via and T by a Relaxed Variable Splitting Method with Application to Multi-scale Curve Classification
Fanghui Xue, Jack Xin
We study sparsification of convolutional neural networks (CNN) by a relaxed variable splitting method of and transformed- (T) penalties, with application t…