Backdoor attacks are serious security threats to machine learning models...
Data-poisoning based backdoor attacks aim to insert backdoor into models...
3D facial avatar reconstruction has been a significant research topic in...
A critical yet frequently overlooked challenge in the field of deepfake
...
Recent studies have demonstrated the susceptibility of deep neural netwo...
Deep neural networks (DNNs) can be manipulated to exhibit specific behav...
Data-free meta-learning (DFML) aims to enable efficient learning of new ...
Deep learning based video frame interpolation (VIF) method, aiming to
sy...
Deepfake detection remains a challenging task due to the difficulty of
g...
Backdoor defense, which aims to detect or mitigate the effect of malicio...
In this paper, we study masked autoencoder (MAE) pretraining on videos f...
Fast adversarial training (FAT) is an efficient method to improve robust...
As a popular paradigm of distributed learning, personalized federated
le...
Adversarial machine learning (AML) studies the adversarial phenomenon of...
Although Deep Neural Networks (DNNs) have been widely applied in various...
The permutation flow shop scheduling (PFSS), aiming at finding the optim...
Deep neural networks (DNNs) have been shown to be vulnerable to adversar...
Adversarial Training (AT) has been demonstrated as one of the most effec...
Recent studies have shown that detectors based on deep models are vulner...
With the thriving of deep learning in processing point cloud data, recen...
To explore the vulnerability of deep neural networks (DNNs), many attack...
Fast adversarial training (FAT) effectively improves the efficiency of
s...
Integer programming (IP) is an important and challenging problem. Approx...
Backdoor learning is an emerging and important topic of studying the
vul...
Deep graph learning has achieved remarkable progresses in both business ...
Adversarial training (AT) is always formulated as a minimax problem, of ...
One-shot talking face generation aims at synthesizing a high-quality tal...
Recent studies have revealed that deep neural networks (DNNs) are vulner...
Object detection has been widely used in many safety-critical tasks, suc...
Adversarial training (AT) has been demonstrated to be effective in impro...
Face recognition has been greatly facilitated by the development of deep...
Adversarial training (AT) has been demonstrated as one of the most promi...
Due to its powerful capability of representation learning and high-effic...
The query-based black-box attacks, which don't require any knowledge abo...
Big progress has been achieved in domain adaptation in decades. Existing...
The existing text-guided image synthesis methods can only produce limite...
To generate "accurate" scene graphs, almost all existing methods predict...
To explore the vulnerability of deep neural networks (DNNs), many attack...
Standard Convolutional Neural Networks (CNNs) can be easily fooled by im...
Recently, backdoor attacks pose a new security threat to the training pr...
In this work, we propose TediGAN, a novel framework for multi-modal imag...
Visual scene graph generation is a challenging task. Previous works have...
Recent GAN-based image inpainting approaches adopt an average strategy t...
With the supervision from source domain only in class-level, existing
un...
Although significant progress achieved, multi-label classification is st...
Deep neural networks (DNNs) have demonstrated their power on many widely...
This work studied the score-based black-box adversarial attack problem, ...
Adversarial examples have been well known as a serious threat to deep ne...
In this work, we study the problem of backdoor attacks, which add a spec...
Adversarial examples have been shown to be the severe threat to deep neu...