activity
20182022
most citedTowards Lightweight Transformer via Group-wise Transformation for Vision-and-Language Tasks

66 citations · 445 across the 42 of their papers we have counts for

collaborators

62 papers

cs.SI20222 cited

Self-supervised Graph Representation Learning for Black Market Account Detection

Zequan Xu, Lianyun Li, Hui Li +3

Nowadays, Multi-purpose Messaging Mobile App (MMMA) has become increasingly prevalent. MMMAs attract fraudsters and some cybercriminals provide support for frauds via black market…

cs.LG202217 cited

Make Sharpness-Aware Minimization Stronger: A Sparsified Perturbation Approach

Peng Mi, Li Shen, Tianhe Ren +4

Deep neural networks often suffer from poor generalization caused by complex and non-convex loss landscapes. One of the popular solutions is Sharpness-Aware Minimization (SAM), whi…

cs.CV2022

ECO-TR: Efficient Correspondences Finding Via Coarse-to-Fine Refinement

Dongli Tan, Jiang-Jiang Liu, Xingyu Chen +5

Modeling sparse and dense image matching within a unified functional correspondence model has recently attracted increasing research interest. However, existing efforts mainly focu…

cs.CV202266 cited

Towards Lightweight Transformer via Group-wise Transformation for Vision-and-Language Tasks

Gen Luo, Yiyi Zhou, Xiaoshuai Sun +5

Despite the exciting performance, Transformer is criticized for its excessive parameters and computation cost. However, compressing Transformer remains as an open problem due to it…

cs.CV2022

Training-free Transformer Architecture Search

Qinqin Zhou, Kekai Sheng, Xiawu Zheng +5

Recently, Vision Transformer (ViT) has achieved remarkable success in several computer vision tasks. The progresses are highly relevant to the architecture design, then it is worth…

cs.CV202214 cited

Boosting Crowd Counting via Multifaceted Attention

Hui Lin, Zhiheng Ma, Rongrong Ji +2

This paper focuses on the challenging crowd counting task. As large-scale variations often exist within crowd images, neither fixed-size convolution kernel of CNN nor fixed-size at…