Hui Liu

Enyan Dai, Tianxiang Zhao, Huaisheng Zhu et al.

Graph Neural Networks (GNNs) have made rapid developments in the recent years. Due to their great ability in modeling graph-structured data, GNNs are vastly used in various applications, including high-stakes scenarios such as financial analysis, traffic predictions, and drug discovery. Despite their great potential in benefiting humans in the real world, recent study shows that GNNs can leak private information, are vulnerable to adversarial attacks, can inherit and magnify societal bias from training data and lack interpretability, which have risk of causing unintentional harm to the users and society. For example, existing works demonstrate that attackers can fool the GNNs to give the outcome they desire with unnoticeable perturbation on training graph. GNNs trained on social networks may embed the discrimination in their decision process, strengthening the undesirable societal bias. Consequently, trustworthy GNNs in various aspects are emerging to prevent the harm from GNN models and increase the users' trust in GNNs. In this paper, we give a comprehensive survey of GNNs in the computational aspects of privacy, robustness, fairness, and explainability. For each aspect, we give the taxonomy of the related methods and formulate the general frameworks for the multiple categories of trustworthy GNNs. We also discuss the future research directions of each aspect and connections between these aspects to help achieve trustworthiness.

Hui Liu, Bo Zhao, Kehuan Zhang et al.

Although deep neural networks (DNNs) have shown impressive performance on many perceptual tasks, they are vulnerable to adversarial examples that are generated by adding slight but maliciously crafted perturbations to benign images. Adversarial detection is an important technique for identifying adversarial examples before they are entered into target DNNs. Previous studies to detect adversarial examples either targeted specific attacks or required expensive computation. How design a lightweight unsupervised detector is still a challenging problem. In this paper, we propose an AutoEncoder-based Adversarial Examples (AEAE) detector, that can guard DNN models by detecting adversarial examples with low computation in an unsupervised manner. The AEAE includes only a shallow autoencoder but plays two roles. First, a well-trained autoencoder has learned the manifold of benign examples. This autoencoder can produce a large reconstruction error for adversarial images with large perturbations, so we can detect significantly perturbed adversarial examples based on the reconstruction error. Second, the autoencoder can filter out the small noise and change the DNN's prediction on adversarial examples with small perturbations. It helps to detect slightly perturbed adversarial examples based on the prediction distance. To cover these two cases, we utilize the reconstruction error and prediction distance from benign images to construct a two-tuple feature set and train an adversarial detector using the isolation forest algorithm. We show empirically that the AEAE is unsupervised and inexpensive against the most state-of-the-art attacks. Through the detection in these two cases, there is nowhere to hide adversarial examples.

Pengfei He, Yue Xing, Juanhui Li et al.

TThis paper argues that \textbf{a comprehensive vulnerability analysis is essential for building trustworthy Large Language Model-based Multi-Agent Systems (LLM-MAS)}. These systems, which consist of multiple LLM-powered agents working collaboratively, are increasingly deployed in high-stakes applications but face novel security threats due to their complex structures. While single-agent vulnerabilities are well-studied, LLM-MAS introduces unique attack surfaces through inter-agent communication, trust relationships, and tool integration that remain significantly underexplored. We present a systematic framework for vulnerability analysis of LLM-MAS that unifies diverse research. For each type of vulnerability, we define formal threat models grounded in practical attacker capabilities and illustrate them using real-world LLM-MAS applications. This formulation enables rigorous quantification of vulnerability across different architectures and provides a foundation for designing meaningful evaluation benchmarks. We also identify critical open challenges: (1) developing benchmarks specifically tailored to LLM-MAS vulnerability assessment, (2) considering new potential attacks specific to multi-agent architectures, and (3) implementing trust management systems that can enforce security in LLM-MAS. This research provides essential groundwork for future efforts to enhance LLM-MAS trustworthiness.

Yingqian Cui, Jie Ren, Han Xu et al.

Recently, Generative Diffusion Models (GDMs) have showcased their remarkable capabilities in learning and generating images. A large community of GDMs has naturally emerged, further promoting the diversified applications of GDMs in various fields. However, this unrestricted proliferation has raised serious concerns about copyright protection. For example, artists including painters and photographers are becoming increasingly concerned that GDMs could effortlessly replicate their unique creative works without authorization. In response to these challenges, we introduce a novel watermarking scheme, DiffusionShield, tailored for GDMs. DiffusionShield protects images from copyright infringement by GDMs through encoding the ownership information into an imperceptible watermark and injecting it into the images. Its watermark can be easily learned by GDMs and will be reproduced in their generated images. By detecting the watermark from generated images, copyright infringement can be exposed with evidence. Benefiting from the uniformity of the watermarks and the joint optimization method, DiffusionShield ensures low distortion of the original image, high watermark detection performance, and the ability to embed lengthy messages. We conduct rigorous and comprehensive experiments to show the effectiveness of DiffusionShield in defending against infringement by GDMs and its superiority over traditional watermarking methods. The code for DiffusionShield is accessible in https://github.com/Yingqiancui/DiffusionShield.

Hui Liu, Bo Zhao, Yuefeng Peng et al.

Deep neural networks (DNNs) are threatened by adversarial examples. Adversarial detection, which distinguishes adversarial images from benign images, is fundamental for robust DNN-based services. Image transformation is one of the most effective approaches to detect adversarial examples. During the last few years, a variety of image transformations have been studied and discussed to design reliable adversarial detectors. In this paper, we systematically synthesize the recent progress on adversarial detection via image transformations with a novel classification method. Then, we conduct extensive experiments to test the detection performance of image transformations against state-of-the-art adversarial attacks. Furthermore, we reveal that each individual transformation is not capable of detecting adversarial examples in a robust way, and propose a DNN-based approach referred to as \emph{AdvJudge}, which combines scores of 9 image transformations. Without knowing which individual scores are misleading or not misleading, AdvJudge can make the right judgment, and achieve a significant improvement in detection rate. Finally, we utilize an explainable AI tool to show the contribution of each image transformation to adversarial detection. Experimental results show that the contribution of image transformations to adversarial detection is significantly different, the combination of them can significantly improve the generic detection ability against state-of-the-art adversarial attacks.

Hui Liu, Bo Zhao, Minzhi Ji et al.

Deep neural networks (DNNs) are under threat from adversarial example attacks. The adversary can easily change the outputs of DNNs by adding small well-designed perturbations to inputs. Adversarial example detection is a fundamental work for robust DNNs-based service. Adversarial examples show the difference between humans and DNNs in image recognition. From a human-centric perspective, image features could be divided into dominant features that are comprehensible to humans, and recessive features that are incomprehensible to humans, yet are exploited by DNNs. In this paper, we reveal that imperceptible adversarial examples are the product of recessive features misleading neural networks, and an adversarial attack is essentially a kind of method to enrich these recessive features in the image. The imperceptibility of the adversarial examples indicates that the perturbations enrich recessive features, yet hardly affect dominant features. Therefore, adversarial examples are sensitive to filtering off recessive features, while benign examples are immune to such operation. Inspired by this idea, we propose a label-only adversarial detection approach that is referred to as feature-filter. Feature-filter utilizes discrete cosine transform to approximately separate recessive features from dominant features, and gets a mutant image that is filtered off recessive features. By only comparing DNN's prediction labels on the input and its mutant, feature-filter can real-time detect imperceptible adversarial examples at high accuracy and few false positives.

Shaofeng Li, Hui Liu, Tian Dong et al.

Natural language processing (NLP) systems have been proven to be vulnerable to backdoor attacks, whereby hidden features (backdoors) are trained into a language model and may only be activated by specific inputs (called triggers), to trick the model into producing unexpected behaviors. In this paper, we create covert and natural triggers for textual backdoor attacks, \textit{hidden backdoors}, where triggers can fool both modern language models and human inspection. We deploy our hidden backdoors through two state-of-the-art trigger embedding methods. The first approach via homograph replacement, embeds the trigger into deep neural networks through the visual spoofing of lookalike character replacement. The second approach uses subtle differences between text generated by language models and real natural text to produce trigger sentences with correct grammar and high fluency. We demonstrate that the proposed hidden backdoors can be effective across three downstream security-critical NLP tasks, representative of modern human-centric NLP systems, including toxic comment detection, neural machine translation (NMT), and question answering (QA). Our two hidden backdoor attacks can achieve an Attack Success Rate (ASR) of at least $97\%$ with an injection rate of only $3\%$ in toxic comment detection, $95.1\%$ ASR in NMT with less than $0.5\%$ injected data, and finally $91.12\%$ ASR against QA updated with only 27 poisoning data samples on a model previously trained with 92,024 samples (0.029\%). We are able to demonstrate the adversary's high success rate of attacks, while maintaining functionality for regular users, with triggers inconspicuous by the human administrators.

Hui Liu, Bo Zhao, Minzhi Ji et al.

Adversarial examples are well-designed input samples, in which perturbations are imperceptible to the human eyes, but easily mislead the output of deep neural networks (DNNs). Existing works synthesize adversarial examples by leveraging simple metrics to penalize perturbations, that lack sufficient consideration of the human visual system (HVS), which produces noticeable artifacts. To explore why the perturbations are visible, this paper summarizes four primary factors affecting the perceptibility of human eyes. Based on this investigation, we design a multi-factor metric MulFactorLoss for measuring the perceptual loss between benign examples and adversarial ones. In order to test the imperceptibility of the multi-factor metric, we propose a novel black-box adversarial attack that is referred to as GreedyFool. GreedyFool applies differential evolution to evaluate the effects of perturbed pixels on the confidence of a target DNN, and introduces greedy approximation to automatically generate adversarial perturbations. We conduct extensive experiments on the ImageNet and CIFRA-10 datasets and a comprehensive user study with 60 participants. The experimental results demonstrate that MulFactorLoss is a more imperceptible metric than the existing pixelwise metrics, and GreedyFool achieves a 100% success rate in a black-box manner.