The integration of Large Language Models (LLMs) into clinical applications presents transformative potential but is undermined by the critical risk of hallucination, the generation of plausible but factually incorrect information. Such failures pose a direct threat to patient safety and the integrity of clinical decision-making. To address this challenge, we introduce MHB, a novel and comprehensive benchmark framework designed to evaluate LLM reliability in two complex, high-stakes clinical contexts: multi-turn medical dialogues and clinical case report analysis. The core of our contribution is a systematic methodology for generating adversarial test cases by injecting ``hallucination traps" into realistic medical data, guided by a fine-grained taxonomy of clinical errors. MHB, comprising 4,695 samples and 20,288 evaluation rubrics, underwent a rigorous, two-stage validation by a panel of 60 licensed physicians from top-tier hospitals, ensuring high clinical realism and consistency. This comprehensive assessment of leading LLMs revealed significant, clinically relevant shortcomings across the board. Even the best-performing model, Claude-4-Sonnet, exhibited a hallucination rate of 29.1%, with some open-source models exceeding 57.0%. All models struggled with specific traps, like fabricated medical data or non-existent guidelines, highlighting prevalent systemic weaknesses.

Machine unlearning is a newly popularized technique for removing specific training data from a trained model, enabling it to comply with data deletion requests. While it protects the rights of users requesting unlearning, it also introduces new privacy risks. Prior works have primarily focused on the privacy of data that has been unlearned, while the risks to retained data remain largely unexplored. To address this gap, we focus on the privacy risks of retained data and, for the first time, reveal the vulnerabilities introduced by machine unlearning under the dual-view setting, where an adversary can query both the original and the unlearned models. From an information-theoretic perspective, we introduce the concept of privacy knowledge gain and demonstrate that the dual-view setting allows adversaries to obtain more information than querying either model alone, thereby amplifying privacy leakage. To effectively demonstrate this threat, we propose DVIA, a Dual-View Inference Attack, which extracts membership information on retained data using black-box queries to both models. DVIA eliminates the need to train an attack model and employs a lightweight likelihood ratio inference module for efficient inference. Experiments across different datasets and model architectures validate the effectiveness of DVIA and highlight the privacy risks inherent in the dual-view setting.

Deep neural networks (DNNs) are susceptible to universal adversarial perturbations (UAPs). These perturbations are meticulously designed to fool the target model universally across all sample classes. Unlike instance-specific adversarial examples (AEs), generating UAPs is more complex because they must be generalized across a wide range of data samples and models. Our research reveals that existing universal attack methods, which optimize UAPs using DNNs with static model parameter snapshots, do not fully leverage the potential of DNNs to generate more effective UAPs. Rather than optimizing UAPs against static DNN models with a fixed training set, we suggest using dynamic model-data pairs to generate UAPs. In particular, we introduce a dynamic maximin optimization strategy, aiming to optimize the UAP across a variety of optimal model-data pairs. We term this approach DM-UAP. DM-UAP utilizes an iterative max-min-min optimization framework that refines the model-data pairs, coupled with a curriculum UAP learning algorithm to examine the combined space of model parameters and data thoroughly. Comprehensive experiments on the ImageNet dataset demonstrate that the proposed DM-UAP markedly enhances both cross-sample universality and cross-model transferability of UAPs. Using only 500 samples for UAP generation, DM-UAP outperforms the state-of-the-art approach with an average increase in fooling ratio of 12.108%.

Direct Prompt Injection (DPI) attacks pose a critical security threat to Large Language Models (LLMs) due to their low barrier of execution and high potential damage.To address the impracticality of existing white-box/gray-box methods and the poor transferability of blackbox methods, we propose an activations-guided prompt injection attack framework.We first construct an Energy-based Model (EBM) using activations from a surrogate model to evaluate the quality of adversarial prompts.Guided by the trained EBM, we employ the tokenlevel Markov Chain Monte Carlo (MCMC) sampling to adaptively optimize adversarial prompts, thereby enabling gradient-free blackbox attacks.Experimental results demonstrate our superior cross-model transferability, achieving 49.6% attack success rate (ASR) across five mainstream LLMs and 34.6% improvement over human-crafted prompts, and maintaining 36.6%ASR on unseen task scenarios.Interpretability analysis reveals a correlation between activations and attack effectiveness, highlighting the critical role of semantic patterns in transferable vulnerability exploitation.

Federated Learning (FL) is a distributed paradigm aimed at protecting participant data privacy by exchanging model parameters to achieve high-quality model training. However, this distributed nature also makes FL highly vulnerable to backdoor attacks. Notably, the recently proposed state-of-the-art (SOTA) attack, 3DFed (SP2023), uses an indicator mechanism to determine whether the backdoor models have been accepted by the defender and adaptively optimizes backdoor models, rendering existing defenses ineffective. In this paper, we first reveal that the failure of existing defenses lies in the employment of empirical statistical measures that are loosely coupled with backdoor attacks. Motivated by this, we propose a Malignity-Aware backdooR defenSe (MARS) that leverages backdoor energy (BE) to indicate the malicious extent of each neuron. To amplify malignity, we further extract the most prominent BE values from each model to form a concentrated backdoor energy (CBE). Finally, a novel Wasserstein distance-based clustering method is introduced to effectively identify backdoor models. Extensive experiments demonstrate that MARS can defend against SOTA backdoor attacks and significantly outperforms existing defenses.

Transfer learning from pre-trained encoders has become essential in modern machine learning, enabling efficient model adaptation across diverse tasks. However, this combination of pre-training and downstream adaptation creates an expanded attack surface, exposing models to sophisticated backdoor embedding at both the encoder and dataset levels—an area often overlooked in prior research. Additionally, the limited computational resources typically available to users of pre-trained encoders constrain the effectiveness of generic backdoor defenses compared to end-to-end training from scratch. In this work, we investigate how to mitigate potential backdoor risks in resource-constrained transfer learning scenarios. Specifically, we first conduct an exhaustive analysis of existing defense strategies, revealing that many follow a reactive workflow based on assumptions that do not scale to unknown threats, novel attack types, or different training paradigms. In response, we introduce a proactive mindset focused on identifying clean elements and propose the Trusted Core (T-Core) Bootstrapping framework, which emphasizes the importance of pinpointing trustworthy data and neurons to enhance model security. Our empirical evaluations demonstrate the effectiveness and superiority of T-Core, specifically assessing 5 encoder poisoning attacks, 7 dataset poisoning attacks, and 14 baseline defenses across 5 benchmark datasets, addressing 4 scenarios of 3 potential backdoor threats.

Traditional unlearnable strategies have been proposed to prevent unauthorized users from training on the 2D image data. With more 3D point cloud data containing sensitivity information, unauthorized usage of this new type data has also become a serious concern. To address this, we propose the first integral unlearnable framework for 3D point clouds including two processes: (i) we propose an unlearnable data protection scheme, involving a class-wise setting established by a category-adaptive allocation strategy and multi-transformations assigned to samples; (ii) we propose a data restoration scheme that utilizes class-wise inverse matrix transformation, thus enabling authorized-only training for unlearnable data. This restoration process is a practical issue overlooked in most existing unlearnable literature, i.e., even authorized users struggle to gain knowledge from 3D unlearnable data. Both theoretical and empirical results (including 6 datasets, 16 models, and 2 tasks) demonstrate the effectiveness of our proposed unlearnable framework. Our code is available at https://github.com/CGCL-codes/UnlearnablePC.

A backdoor attack in deep learning inserts a hidden backdoor in the model to trigger malicious behavior upon specific input patterns. Existing detection approaches assume a metric space (for either the original inputs or their latent representations) in which normal samples and malicious samples are separable. We show that this assumption has a severe limitation by introducing a novel SSDT (Source-Specific and Dynamic-Triggers) backdoor, which obscures the difference between normal samples and malicious samples.To overcome this limitation, we move beyond looking for a perfect metric space that would work for different deep-learning models, and instead resort to more robust topological constructs. We propose TED (Topological Evolution Dynamics) as a model-agnostic basis for robust backdoor detection. The main idea of TED is to view a deep-learning model as a dynamical system that evolves inputs to outputs. In such a dynamical system, a benign input follows a natural evolution trajectory similar to other benign inputs. In contrast, a malicious sample displays a distinct trajectory, since it starts close to benign samples but eventually shifts towards the neighborhood of attacker-specified target samples to activate the backdoor.Extensive evaluations are conducted on vision and natural language datasets across different network architectures. The results demonstrate that TED not only achieves a high detection rate, but also significantly outperforms existing state-of-the-art detection approaches, particularly in addressing the sophisticated SSDT attack. The code to reproduce the results is made public on GitHub.

Adversarial examples for deep neural networks (DNNs) are transferable: examples that successfully fool one white-box surrogate model can also deceive other black-box models with different architectures. Although a bunch of empirical studies have provided guidance on generating highly transferable adversarial examples, many of these findings fail to be well explained and even lead to confusing or inconsistent advice for practical use.In this paper, we take a further step towards understanding adversarial transferability, with a particular focus on surrogate aspects. Starting from the intriguing "little robustness" phenomenon, where models adversarially trained with mildly perturbed adversarial samples can serve as better surrogates for transfer attacks, we attribute it to a trade-off between two dominant factors: model smoothness and gradient similarity. Our research focuses on their joint effects on transferability, rather than demonstrating the separate relationships alone. Through a combination of theoretical and empirical analyses, we hypothesize that the data distribution shift induced by off-manifold samples in adversarial training is the reason that impairs gradient similarity.Building on these insights, we further explore the impacts of prevalent data augmentation and gradient regularization on transferability and analyze how the trade-off manifests in various training methods, thus building a comprehensive blueprint for the regulation mechanisms behind transferability. Finally, we provide a general route for constructing superior surrogates to boost transferability, which optimizes both model smoothness and gradient similarity simultaneously, e.g., the combination of input gradient regularization and sharpness-aware minimization (SAM), validated by extensive experiments. In summary, we call for attention to the united impacts of these two factors for launching effective transfer attacks, rather than optimizing one while ignoring the other, and emphasize the crucial role of manipulating surrogate models.

With the evolution of self-supervised learning, the pre-training paradigm has emerged as a predominant solution within the deep learning landscape. Model providers furnish pre-trained encoders designed to function as versatile feature extractors, enabling downstream users to harness the benefits of expansive models with minimal effort through fine-tuning. Nevertheless, recent works have exposed a vulnerability in pre-trained encoders, highlighting their susceptibility to downstream-agnostic adversarial examples (DAEs) meticulously crafted by attackers. The lingering question pertains to the feasibility of fortifying the robustness of downstream models against DAEs, particularly in scenarios where the pre-trained encoders are publicly accessible to the attackers.In this paper, we initially delve into existing defensive mechanisms against adversarial examples within the pre-training paradigm. Our findings reveal that the failure of current defenses stems from the domain shift between pre-training data and downstream tasks, as well as the sensitivity of encoder parameters. In response to these challenges, we propose Genetic Evolution-Nurtured Adversarial Fine-tuning (Gen-AF), a two-stage adversarial fine-tuning approach aimed at enhancing the robustness of downstream models. Gen-AF employs a genetic-directed dual-track adversarial fine-tuning strategy in its first stage to effectively inherit the pre-trained encoder. This involves optimizing the pre-trained encoder and classifier separately while incorporating genetic regularization to preserve the model’s topology. In the second stage, Gen-AF assesses the robust sensitivity of each layer and creates a dictionary, based on which the top-k robust redundant layers are selected with the remaining layers held fixed. Upon this foundation, we conduct evolutionary adaptability fine-tuning to further enhance the model’s generalizability. Our extensive experiments, conducted across ten self-supervised training methods and six datasets, demonstrate that Gen-AF attains high testing accuracy and robust testing accuracy against state-of-the-art DAEs.

Backdoor attacks for point clouds have elicited mounting interest with the proliferation of deep learning. The point cloud classifiers can be vulnerable to malicious actors who seek to manipulate or fool the model with specific backdoor triggers. Detecting and rejecting backdoor samples during the inference stage can effectively alleviate backdoor attacks. Recently, some black-box test-time backdoor sample detection methods have been proposed in the 2D image domain, without any underlying assumptions about the backdoor triggers. However, upon examination, we have found that these detection techniques are not effective for 3D point clouds. As a result, there is a pressing need to bridge the gap for the development of a universal approach that is specifically designed for 3D point clouds.

Multimodal contrastive learning aims to train a general-purpose feature extractor, such as CLIP, on vast amounts of raw, unlabeled paired image-text data. This can greatly benefit various complex downstream tasks, including cross-modal image-text retrieval and image classification. Despite its promising prospect, the security issue of cross-modal pre-trained encoder has not been fully explored yet, especially when the pre-trained encoder is publicly available for commercial use.

While deep face recognition (FR) systems have shown amazing performance in identification and verification, they also arouse privacy concerns for their excessive surveillance on users, especially for public face images widely spread on social networks. Recently, some studies adopt adversarial examples to protect photos from being identified by unauthorized face recognition systems. However, existing methods of generating adversarial face images suffer from many limitations, such as awkward visual, white-box setting, weak transferability, making them difficult to be applied to protect face privacy in reality. In this paper, we propose adversarial makeup transfer GAN (AMT-GAN)11https://github.com/CGCL-codes/AMT-GAN, a novel face protection method aiming at constructing adversarial face images that preserve stronger black-box transferability and better visual quality simultaneously. AMT-GAN leverages generative adversarial networks (GAN) to synthesize adversarial face images with makeup transferred from reference images. In particular, we introduce a new regularization module along with a joint training strategy to reconcile the conflicts between the adversarial noises and the cycle consistence loss in makeup transfer, achieving a desirable balance between the attack strength and visual changes. Extensive experiments verify that compared with state of the arts, AMT-GAN can not only preserve a comfortable visual quality, but also achieve a higher attack success rate over commercial FR APIs, including Face++, Aliyun, and Microsoft.

Due to its powerful feature learning capability and high efficiency, deep hashing has achieved great success in large-scale image retrieval. Meanwhile, extensive works have demonstrated that deep neural networks (DNNs) are susceptible to adversarial examples, and exploring adversarial attack against deep hashing has attracted many research efforts. Nevertheless, backdoor attack, another famous threat to DNNs, has not been studied for deep hashing yet. Although various backdoor attacks have been proposed in the field of image classification, existing approaches failed to realize a truly imperceptive backdoor attack that enjoys invisible triggers and clean label setting simultaneously, and they cannot meet the intrinsic demand of image retrieval backdoor.

In this paper, we propose AdvHash, the first targeted mismatch attack on deep hashing through adversarial patch. After superimposed with the same adversarial patch, any query image with a chosen label will retrieve a set of irrelevant images with the target label. Concretely, we first formulate a set-to-set problem, where a set of samples are pushed into a predefined clustered area in the Hamming space. Then we obtain a target anchor hash code and transform the attack to a set-to-point optimization. In order to generate a image-agnostic stable adversarial patch for a chosen label more efficiently, we propose a product-based weighted gradient aggregation strategy to dynamically adjust the gradient directions of the patch, by exploiting the Hamming distances between training samples and the target anchor hash code and assigning different weights to discriminatively aggregate gradients. Extensive experiments on benchmark datasets verify that AdvHash is highly effective at attacking two state-of-the-art deep hashing schemes. Our codes are available at: https://github.com/CGCL-codes/AdvHash.