Xingjun Ma

Jiaming Zhang, Xingjun Ma, Qi Yi et al.

There is a growing interest in developing unlearnable examples (UEs) against visual privacy leaks on the Internet. UEs are training samples added with invisible but unlearnable noise, which have been found can prevent unauthorized training of machine learning models. UEs typically are generated via a bilevel optimization framework with a surrogate model to remove (minimize) errors from the original samples, and then applied to protect the data against unknown target models. However, existing UE generation methods all rely on an ideal assumption called label-consistency, where the hackers and protectors are assumed to hold the same label for a given sample. In this work, we propose and promote a more practical label-agnostic setting, where the hackers may exploit the protected data quite differently from the protectors. E.g., a m-class unlearnable dataset held by the protector may be exploited by the hacker as a n-class dataset. Existing UE generation methods are rendered ineffective in this challenging setting. To tackle this challenge, we present a novel technique called Unlearnable Clusters (UCs) to generate label-agnostic unlearnable examples with cluster-wise perturbations. Furthermore, we propose to leverage VisionandLanguage Pre-trained Models (VLPMs) like CLIP as the surrogate model to improve the transferability of the crafted UCs to diverse domains. We empirically verify the effectiveness of our proposed approach under a variety of settings with different datasets, target models, and even commercial platforms Microsoft Azure and Baidu PaddlePaddle. Code is available at \url{https://github.com/jiamingzhang94/Unlearnable-Clusters}.

Hanxun Huang, Xingjun Ma, Sarah Erfani et al.

This paper proposes a simple method to distill and detect backdoor patterns within an image: \emph{Cognitive Distillation} (CD). The idea is to extract the "minimal essence" from an input image responsible for the model's prediction. CD optimizes an input mask to extract a small pattern from the input image that can lead to the same model output (i.e., logits or deep features). The extracted pattern can help understand the cognitive mechanism of a model on clean vs. backdoor images and is thus called a \emph{Cognitive Pattern} (CP). Using CD and the distilled CPs, we uncover an interesting phenomenon of backdoor attacks: despite the various forms and sizes of trigger patterns used by different attacks, the CPs of backdoor samples are all surprisingly and suspiciously small. One thus can leverage the learned mask to detect and remove backdoor examples from poisoned training datasets. We conduct extensive experiments to show that CD can robustly detect a wide range of advanced backdoor attacks. We also show that CD can potentially be applied to help detect potential biases from face datasets. Code is available at \url{https://github.com/HanxunH/CognitiveDistillation}.

Yixu Wang, Yan Teng, Kexin Huang et al.

The growing awareness of safety concerns in large language models (LLMs) has sparked considerable interest in the evaluation of safety. This study investigates an under-explored issue about the evaluation of LLMs, namely the substantial discrepancy in performance between multiple-choice questions and open-ended questions. Inspired by research on jailbreak attack patterns, we argue this is caused by mismatched generalization. That is, LLM only remembers the answer style for open-ended safety questions, which makes it unable to solve other forms of safety tests. We refer to this phenomenon as fake alignment and construct a comparative benchmark to empirically verify its existence in LLMs. We introduce a Fake alIgNment Evaluation (FINE) framework and two novel metrics--Consistency Score (CS) and Consistent Safety Score (CSS), which jointly assess two complementary forms of evaluation to quantify fake alignment and obtain corrected performance estimation. Applying FINE to 14 widely-used LLMs reveals several models with purported safety are poorly aligned in practice. Subsequently, we found that multiple-choice format data can also be used as high-quality contrast distillation-based fine-tuning data, which can strongly improve the alignment consistency of LLMs with minimal fine-tuning overhead. For data and code, see https://github.com/AIFlames/Fake-Alignment.

Zhiyuan Zhang, Lingjuan Lyu, Xingjun Ma et al. · pku

Deep Neural Networks (DNNs) are known to be vulnerable to backdoor attacks. In Natural Language Processing (NLP), DNNs are often backdoored during the fine-tuning process of a large-scale Pre-trained Language Model (PLM) with poisoned samples. Although the clean weights of PLMs are readily available, existing methods have ignored this information in defending NLP models against backdoor attacks. In this work, we take the first step to exploit the pre-trained (unfine-tuned) weights to mitigate backdoors in fine-tuned language models. Specifically, we leverage the clean pre-trained weights via two complementary techniques: (1) a two-step Fine-mixing technique, which first mixes the backdoored weights (fine-tuned on poisoned data) with the pre-trained weights, then fine-tunes the mixed weights on a small subset of clean data; (2) an Embedding Purification (E-PUR) technique, which mitigates potential backdoors existing in the word embeddings. We compare Fine-mixing with typical backdoor mitigation methods on three single-sentence sentiment classification tasks and two sentence-pair classification tasks and show that it outperforms the baselines by a considerable margin in all scenarios. We also show that our E-PUR method can benefit existing mitigation methods. Our work establishes a simple but strong baseline defense for secure fine-tuned NLP models against backdoor attacks.

Jiaming Zhang, Xingjun Ma, Xin Wang et al.

With the rapid advancement of multimodal learning, pre-trained Vision-Language Models (VLMs) such as CLIP have demonstrated remarkable capacities in bridging the gap between visual and language modalities. However, these models remain vulnerable to adversarial attacks, particularly in the image modality, presenting considerable security risks. This paper introduces Adversarial Prompt Tuning (AdvPT), a novel technique to enhance the adversarial robustness of image encoders in VLMs. AdvPT innovatively leverages learnable text prompts and aligns them with adversarial image embeddings, to address the vulnerabilities inherent in VLMs without the need for extensive parameter training or modification of the model architecture. We demonstrate that AdvPT improves resistance against white-box and black-box adversarial attacks and exhibits a synergistic effect when combined with existing image-processing-based defense techniques, further boosting defensive capabilities. Comprehensive experimental analyses provide insights into adversarial prompt tuning, a novel paradigm devoted to improving resistance to adversarial images through textual input modifications, paving the way for future robust multimodal learning research. These findings open up new possibilities for enhancing the security of VLMs. Our code is available at https://github.com/jiamingzhang94/Adversarial-Prompt-Tuning.

Xiangshan Gao, Xingjun Ma, Jingyi Wang et al.

Federated learning (FL) is a collaborative learning paradigm where participants jointly train a powerful model without sharing their private data. One desirable property for FL is the implementation of the right to be forgotten (RTBF), i.e., a leaving participant has the right to request to delete its private data from the global model. However, unlearning itself may not be enough to implement RTBF unless the unlearning effect can be independently verified, an important aspect that has been overlooked in the current literature. In this paper, we prompt the concept of verifiable federated unlearning, and propose VeriFi, a unified framework integrating federated unlearning and verification that allows systematic analysis of the unlearning and quantification of its effect, with different combinations of multiple unlearning and verification methods. In VeriFi, the leaving participant is granted the right to verify (RTV), that is, the participant notifies the server before leaving, then actively verifies the unlearning effect in the next few communication rounds. The unlearning is done at the server side immediately after receiving the leaving notification, while the verification is done locally by the leaving participant via two steps: marking (injecting carefully-designed markers to fingerprint the leaver) and checking (examining the change of the global model's performance on the markers). Based on VeriFi, we conduct the first systematic and large-scale study for verifiable federated unlearning, considering 7 unlearning methods and 5 verification methods. Particularly, we propose a more efficient and FL-friendly unlearning method, and two more effective and robust non-invasive-verification methods. We extensively evaluate VeriFi on 7 datasets and 4 types of deep learning models. Our analysis establishes important empirical understandings for more trustworthy federated unlearning.

Hanxun Huang, Qizhou Wang, Xingjun Ma et al.

Audio self-supervised learning (SSL) aims to learn general-purpose representations from large-scale unlabeled audio data. While recent advances have been driven mainly by generative reconstruction objectives, contrastive approaches remain less explored, partly due to the difficulty of designing effective audio augmentations and the large batch sizes required for contrastive pre-training. We introduce \textbf{AudioMosaic}, a contrastive learning-based audio encoder for general audio understanding. During pre-training, AudioMosaic constructs positive pairs by applying structured time-frequency masking to spectrogram patches, which reduces memory usage and enables efficient large-batch training. Compared with generative approaches, the AudioMosaic encoder learns more discriminative utterance-level representations that demonstrate strong transferability across datasets, domains, and acoustic conditions. Extensive experiments show that AudioMosaic achieves state-of-the-art performance on several standard audio benchmarks under both linear probing and fine-tuning. We further show that integrating the pretrained AudioMosaic encoder into audio-language models improves performance on audio-language tasks. The code is publicly available in our \href{https://github.com/HanxunH/AudioMosaic}{GitHub repository}.

Xin Wang, Kai Chen, Xingjun Ma et al.

Deep neural networks (DNNs) are known to be vulnerable to adversarial attacks even under a black-box setting where the adversary can only query the model. Particularly, query-based black-box adversarial attacks estimate adversarial gradients based on the returned probability vectors of the target model for a sequence of queries. During this process, the queries made to the target model are intermediate adversarial examples crafted at the previous attack step, which share high similarities in the pixel space. Motivated by this observation, stateful detection methods have been proposed to detect and reject query-based attacks. While demonstrating promising results, these methods either have been evaded by more advanced attacks or suffer from low efficiency in terms of the number of shots (queries) required to detect different attacks. Arguably, the key challenge here is to assign high similarity scores for any two intermediate adversarial examples perturbed from the same clean image. To address this challenge, we propose a novel Adversarial Contrastive Prompt Tuning (ACPT) method to robustly fine-tune the CLIP image encoder to extract similar embeddings for any two intermediate adversarial queries. With ACPT, we further introduce a detection framework AdvQDet that can detect 7 state-of-the-art query-based attacks with $>99\%$ detection rate within 5 shots. We also show that ACPT is robust to 3 types of adaptive attacks. Code is available at https://github.com/xinwong/AdvQDet.

Yige Li, Hanxun Huang, Yunhan Zhao et al.

Generative large language models (LLMs) have achieved state-of-the-art results on a wide range of tasks, yet they remain susceptible to backdoor attacks: carefully crafted triggers in the input can manipulate the model to produce adversary-specified outputs. While prior research has predominantly focused on backdoor risks in vision and classification settings, the vulnerability of LLMs in open-ended text generation remains underexplored. To fill this gap, we introduce BackdoorLLM (Our BackdoorLLM benchmark was awarded First Prize in the SafetyBench competition, https://www.mlsafety.org/safebench/winners, organized by the Center for AI Safety, https://safe.ai/.), the first comprehensive benchmark for systematically evaluating backdoor threats in text-generation LLMs. BackdoorLLM provides: (i) a unified repository of benchmarks with a standardized training and evaluation pipeline; (ii) a diverse suite of attack modalities, including data poisoning, weight poisoning, hidden-state manipulation, and chain-of-thought hijacking; (iii) over 200 experiments spanning 8 distinct attack strategies, 7 real-world scenarios, and 6 model architectures; (iv) key insights into the factors that govern backdoor effectiveness and failure modes in LLMs; and (v) a defense toolkit encompassing 7 representative mitigation techniques. Our code and datasets are available at https://github.com/bboylyg/BackdoorLLM. We will continuously incorporate emerging attack and defense methodologies to support the research in advancing the safety and reliability of LLMs.

Jie Ren, Han Xu, Yuxuan Wan et al.

With more people publishing their personal data online, unauthorized data usage has become a serious concern. The unlearnable strategies have been introduced to prevent third parties from training on the data without permission. They add perturbations to the users' data before publishing, which aims to make the models trained on the perturbed published dataset invalidated. These perturbations have been generated for a specific training setting and a target dataset. However, their unlearnable effects significantly decrease when used in other training settings and datasets. To tackle this issue, we propose a novel unlearnable strategy based on Classwise Separability Discriminant (CSD), which aims to better transfer the unlearnable effects to other training settings and datasets by enhancing the linear separability. Extensive experiments demonstrate the transferability of the proposed unlearnable examples across training settings and datasets.

Yilun Zhang, Yuqian Fu, Xingjun Ma et al.

Deep learning has achieved great success in video recognition, yet still struggles to recognize novel actions when faced with only a few examples. To tackle this challenge, few-shot action recognition methods have been proposed to transfer knowledge from a source dataset to a novel target dataset with only one or a few labeled videos. However, existing methods mainly focus on modeling the temporal relations between the query and support videos while ignoring the spatial relations. In this paper, we find that the spatial misalignment between objects also occurs in videos, notably more common than the temporal inconsistency. We are thus motivated to investigate the importance of spatial relations and propose a more accurate few-shot action recognition method that leverages both spatial and temporal information. Particularly, a novel Spatial Alignment Cross Transformer (SA-CT) which learns to re-adjust the spatial relations and incorporates the temporal information is contributed. Experiments reveal that, even without using any temporal information, the performance of SA-CT is comparable to temporal based methods on 3/4 benchmarks. To further incorporate the temporal information, we propose a simple yet effective Temporal Mixer module. The Temporal Mixer enhances the video representation and improves the performance of the full SA-CT model, achieving very competitive results. In this work, we also exploit large-scale pretrained models for few-shot action recognition, providing useful insights for this research direction.

Xiang Zheng, Xingjun Ma, Chao Shen et al.

This paper investigates two fundamental problems that arise when utilizing Intrinsic Motivation (IM) for reinforcement learning in Reward-Free Pre-Training (RFPT) tasks and Exploration with Intrinsic Motivation (EIM) tasks: 1) how to design an effective intrinsic objective in RFPT tasks, and 2) how to reduce the bias introduced by the intrinsic objective in EIM tasks. Existing IM methods suffer from static skills, limited state coverage, sample inefficiency in RFPT tasks, and suboptimality in EIM tasks. To tackle these problems, we propose \emph{Constrained Intrinsic Motivation (CIM)} for RFPT and EIM tasks, respectively: 1) CIM for RFPT maximizes the lower bound of the conditional state entropy subject to an alignment constraint on the state encoder network for efficient dynamic and diverse skill discovery and state coverage maximization; 2) CIM for EIM leverages constrained policy optimization to adaptively adjust the coefficient of the intrinsic objective to mitigate the distraction from the intrinsic objective. In various MuJoCo robotics environments, we empirically show that CIM for RFPT greatly surpasses fifteen IM methods for unsupervised skill discovery in terms of skill diversity, state coverage, and fine-tuning performance. Additionally, we showcase the effectiveness of CIM for EIM in redeeming intrinsic rewards when task rewards are exposed from the beginning. Our code is available at https://github.com/x-zheng16/CIM.

Chen Chen, Yuchen Liu, Xingjun Ma et al.

Recent studies have shown that, like traditional machine learning, federated learning (FL) is also vulnerable to adversarial attacks. To improve the adversarial robustness of FL, federated adversarial training (FAT) methods have been proposed to apply adversarial training locally before global aggregation. Although these methods demonstrate promising results on independent identically distributed (IID) data, they suffer from training instability on non-IID data with label skewness, resulting in degraded natural accuracy. This tends to hinder the application of FAT in real-world applications where the label distribution across the clients is often skewed. In this paper, we study the problem of FAT under label skewness, and reveal one root cause of the training instability and natural accuracy degradation issues: skewed labels lead to non-identical class probabilities and heterogeneous local models. We then propose a Calibrated FAT (CalFAT) approach to tackle the instability issue by calibrating the logits adaptively to balance the classes. We show both theoretically and empirically that the optimization of CalFAT leads to homogeneous local models across the clients and better convergence points.

Yuhua Sun, Tailai Zhang, Xingjun Ma et al.

Crowd counting is a regression task that estimates the number of people in a scene image, which plays a vital role in a range of safety-critical applications, such as video surveillance, traffic monitoring and flow control. In this paper, we investigate the vulnerability of deep learning based crowd counting models to backdoor attacks, a major security threat to deep learning. A backdoor attack implants a backdoor trigger into a target model via data poisoning so as to control the model's predictions at test time. Different from image classification models on which most of existing backdoor attacks have been developed and tested, crowd counting models are regression models that output multi-dimensional density maps, thus requiring different techniques to manipulate. In this paper, we propose two novel Density Manipulation Backdoor Attacks (DMBA$^{-}$ and DMBA$^{+}$) to attack the model to produce arbitrarily large or small density estimations. Experimental results demonstrate the effectiveness of our DMBA attacks on five classic crowd counting models and four types of datasets. We also provide an in-depth analysis of the unique challenges of backdooring crowd counting models and reveal two key elements of effective attacks: 1) full and dense triggers and 2) manipulation of the ground truth counts or density maps. Our work could help evaluate the vulnerability of crowd counting models to potential backdoor attacks.

Yujing Jiang, Xingjun Ma, Sarah Monazam Erfani et al.

Backdoor attacks have emerged as one of the major security threats to deep learning models as they can easily control the model's test-time predictions by pre-injecting a backdoor trigger into the model at training time. While backdoor attacks have been extensively studied on images, few works have investigated the threat of backdoor attacks on time series data. To fill this gap, in this paper we present a novel generative approach for time series backdoor attacks against deep learning based time series classifiers. Backdoor attacks have two main goals: high stealthiness and high attack success rate. We find that, compared to images, it can be more challenging to achieve the two goals on time series. This is because time series have fewer input dimensions and lower degrees of freedom, making it hard to achieve a high attack success rate without compromising stealthiness. Our generative approach addresses this challenge by generating trigger patterns that are as realistic as real-time series patterns while achieving a high attack success rate without causing a significant drop in clean accuracy. We also show that our proposed attack is resistant to potential backdoor defenses. Furthermore, we propose a novel universal generator that can poison any type of time series with a single generator that allows universal attacks without the need to fine-tune the generative model for new time series datasets.

Jiabin Hua, Hengyuan Xu, Aojie Li et al.

Fine-grained facial expression editing has long been limited by intrinsic semantic overlap. To address this, we construct the Flex Facial Expression (FFE) dataset with continuous affective annotations and establish FFE-Bench to evaluate structural confusion, editing accuracy, linear controllability, and the trade-off between expression editing and identity preservation. We propose PixelSmile, a diffusion framework that disentangles expression semantics via fully symmetric joint training. PixelSmile combines intensity supervision with contrastive learning to produce stronger and more distinguishable expressions, achieving precise and stable linear expression control through textual latent interpolation. Extensive experiments demonstrate that PixelSmile achieves superior disentanglement and robust identity preservation, confirming its effectiveness for continuous, controllable, and fine-grained expression editing, while naturally supporting smooth expression blending.

Weijie Zheng, Xingjun Ma, Hanxun Huang et al.

With the advancement of vision transformers (ViTs) and self-supervised learning (SSL) techniques, pre-trained large ViTs have become the new foundation models for computer vision applications. However, studies have shown that, like convolutional neural networks (CNNs), ViTs are also susceptible to adversarial attacks, where subtle perturbations in the input can fool the model into making false predictions. This paper studies the transferability of such an adversarial vulnerability from a pre-trained ViT model to downstream tasks. We focus on \emph{sample-wise} transfer attacks and propose a novel attack method termed \emph{Downstream Transfer Attack (DTA)}. For a given test image, DTA leverages a pre-trained ViT model to craft the adversarial example and then applies the adversarial example to attack a fine-tuned version of the model on a downstream dataset. During the attack, DTA identifies and exploits the most vulnerable layers of the pre-trained model guided by a cosine similarity loss to craft highly transferable attacks. Through extensive experiments with pre-trained ViTs by 3 distinct pre-training methods, 3 fine-tuning schemes, and across 10 diverse downstream datasets, we show that DTA achieves an average attack success rate (ASR) exceeding 90\%, surpassing existing methods by a huge margin. When used with adversarial training, the adversarial examples generated by our DTA can significantly improve the model's robustness to different downstream transfer attacks.

Xiang Zheng, Xingjun Ma, Cong Wang

Intrinsic motivation is a promising exploration technique for solving reinforcement learning tasks with sparse or absent extrinsic rewards. There exist two technical challenges in implementing intrinsic motivation: 1) how to design a proper intrinsic objective to facilitate efficient exploration; and 2) how to combine the intrinsic objective with the extrinsic objective to help find better solutions. In the current literature, the intrinsic objectives are all designed in a task-agnostic manner and combined with the extrinsic objective via simple addition (or used by itself for reward-free pre-training). In this work, we show that these designs would fail in typical sparse-reward continuous control tasks. To address the problem, we propose Constrained Intrinsic Motivation (CIM) to leverage readily attainable task priors to construct a constrained intrinsic objective, and at the same time, exploit the Lagrangian method to adaptively balance the intrinsic and extrinsic objectives via a simultaneous-maximization framework. We empirically show, on multiple sparse-reward continuous control tasks, that our CIM approach achieves greatly improved performance and sample efficiency over state-of-the-art methods. Moreover, the key techniques of our CIM can also be plugged into existing methods to boost their performances.

Tiehua Zhang, Yuze Liu, Zhishu Shen et al.

Graph neural network (GNN) has gained increasing popularity in recent years owing to its capability and flexibility in modeling complex graph structure data. Among all graph learning methods, hypergraph learning is a technique for exploring the implicit higher-order correlations when training the embedding space of the graph. In this paper, we propose a hypergraph learning framework named LFH that is capable of dynamic hyperedge construction and attentive embedding update utilizing the heterogeneity attributes of the graph. Specifically, in our framework, the high-quality features are first generated by the pairwise fusion strategy that utilizes explicit graph structure information when generating initial node embedding. Afterwards, a hypergraph is constructed through the dynamic grouping of implicit hyperedges, followed by the type-specific hypergraph learning process. To evaluate the effectiveness of our proposed framework, we conduct comprehensive experiments on several popular datasets with eleven state-of-the-art models on both node classification and link prediction tasks, which fall into categories of homogeneous pairwise graph learning, heterogeneous pairwise graph learning, and hypergraph learning. The experiment results demonstrate a significant performance gain (average 12.5% in node classification and 13.3% in link prediction) compared with recent state-of-the-art methods.

Zonghuan Xu, Xingjun Ma

A central challenge in continual learning is forgetting, the loss of performance on previously learned tasks induced by sequential adaptation to new ones. While forgetting has been extensively studied empirically, rigorous theoretical characterizations remain limited. A notable step in this direction is \citet{evron2022catastrophic}, which analyzes forgetting under random orderings of a fixed task collection in overparameterized linear regression. We shift the perspective from order to distribution. Rather than asking how a fixed task collection behaves under random orderings, we study an exact-fit linear regime in which tasks are sampled i.i.d.\ from a task distribution~$Π$, and ask how the generating distribution itself governs forgetting. In this setting, we derive an exact operator identity for the forgetting quantity, revealing a recursive spectral structure. Building on this identity, we establish an unconditional upper bound, identify the leading asymptotic term, and, in generic nondegenerate cases, characterize the convergence rate up to constants. We further relate this rate to geometric properties of the task distribution, clarifying what drives slow or fast forgetting in this model.

Xueqi Ma, Xingjun Ma, Sarah Monazam Erfani et al.

Developing open-set classification methods capable of classifying in-distribution (ID) data while detecting out-of-distribution (OOD) samples is essential for deploying graph neural networks (GNNs) in open-world scenarios. Existing methods typically treat all OOD samples as a single class, despite real-world applications, especially high-stake settings such as fraud detection and medical diagnosis, demanding deeper insights into OOD samples, including their probable labels. This raises a critical question: can OOD detection be extended to OOD classification without true label information? To address this question, we propose a Coarse-to-Fine open-set Classification (CFC) framework that leverages large language models (LLMs) for graph datasets. CFC consists of three key components: a coarse classifier that uses LLM prompts for OOD detection and outlier label generation, a GNN-based fine classifier trained with OOD samples identified by the coarse classifier for enhanced OOD detection and ID classification, and refined OOD classification achieved through LLM prompts and post-processed OOD labels. Unlike methods that rely on synthetic or auxiliary OOD samples, CFC employs semantic OOD instances that are genuinely out-of-distribution based on their inherent meaning, improving interpretability and practical utility. Experimental results show that CFC improves OOD detection by ten percent over state-of-the-art methods on graph and text domains and achieves up to seventy percent accuracy in OOD classification on graph datasets.

Jiahao Zhang, Zilong Wang, Ruofan Wang et al.

As Large Language Models (LLMs) are increasingly being deployed in safety-critical applications, their vulnerability to potential jailbreaks -- malicious prompts that can disable the safety mechanism of LLMs -- has attracted growing research attention. While alignment methods have been proposed to protect LLMs from jailbreaks, many have found that aligned LLMs can still be jailbroken by carefully crafted malicious prompts, producing content that violates policy regulations. Existing jailbreak attacks on LLMs can be categorized into prompt-level methods which make up stories/logic to circumvent safety alignment and token-level attack methods which leverage gradient methods to find adversarial tokens. In this work, we introduce the concept of Ensemble Jailbreak and explore methods that can integrate prompt-level and token-level jailbreak into a more powerful hybrid jailbreak attack. Specifically, we propose a novel EnJa attack to hide harmful instructions using prompt-level jailbreak, boost the attack success rate using a gradient-based attack, and connect the two types of jailbreak attacks via a template-based connector. We evaluate the effectiveness of EnJa on several aligned models and show that it achieves a state-of-the-art attack success rate with fewer queries and is much stronger than any individual jailbreak.

Yunhan Zhao, Xiang Zheng, Lin Luo et al.

In this paper, we focus on black-box defense for VLMs against jailbreak attacks. Existing black-box defense methods are either unimodal or bimodal. Unimodal methods enhance either the vision or language module of the VLM, while bimodal methods robustify the model through text-image representation realignment. However, these methods suffer from two limitations: 1) they fail to fully exploit the cross-modal information, or 2) they degrade the model performance on benign inputs. To address these limitations, we propose a novel blue-team method BlueSuffix that defends target VLMs against jailbreak attacks without compromising its performance under black-box setting. BlueSuffix includes three key components: 1) a visual purifier against jailbreak images, 2) a textual purifier against jailbreak texts, and 3) a blue-team suffix generator using reinforcement fine-tuning for enhancing cross-modal robustness. We empirically show on four VLMs (LLaVA, MiniGPT-4, InstructionBLIP, and Gemini) and four safety benchmarks (Harmful Instruction, AdvBench, MM-SafetyBench, and RedTeam-2K) that BlueSuffix outperforms the baseline defenses by a significant margin. Our BlueSuffix opens up a promising direction for defending VLMs against jailbreak attacks. Code is available at https://github.com/Vinsonzyh/BlueSuffix.

Yige Li, Jiabo He, Hanxun Huang et al.

Backdoor attacks have become a significant threat to the pre-training and deployment of deep neural networks (DNNs). Although numerous methods for detecting and mitigating backdoor attacks have been proposed, most rely on identifying and eliminating the ``shortcut" created by the backdoor, which links a specific source class to a target class. However, these approaches can be easily circumvented by designing multiple backdoor triggers that create shortcuts everywhere and therefore nowhere specific. In this study, we explore the concept of Multi-Trigger Backdoor Attacks (MTBAs), where multiple adversaries leverage different types of triggers to poison the same dataset. By proposing and investigating three types of multi-trigger attacks including \textit{parallel}, \textit{sequential}, and \textit{hybrid} attacks, we demonstrate that 1) multiple triggers can coexist, overwrite, or cross-activate one another, and 2) MTBAs easily break the prevalent shortcut assumption underlying most existing backdoor detection/removal methods, rendering them ineffective. Given the security risk posed by MTBAs, we have created a multi-trigger backdoor poisoning dataset to facilitate future research on detecting and mitigating these attacks, and we also discuss potential defense strategies against MTBAs. Our code is available at https://github.com/bboylyg/Multi-Trigger-Backdoor-Attacks.

Jiayu Li, Yunhan Zhao, Xiang Zheng et al.

Vision-Language-Action (VLA) models enable robots to interpret natural-language instructions and perform diverse tasks, yet their integration of perception, language, and control introduces new safety vulnerabilities. Despite growing interest in attacking such models, the effectiveness of existing techniques remains unclear due to the absence of a unified evaluation framework. One major issue is that differences in action tokenizers across VLA architectures hinder reproducibility and fair comparison. More importantly, most existing attacks have not been validated in real-world scenarios. To address these challenges, we propose AttackVLA, a unified framework that aligns with the VLA development lifecycle, covering data construction, model training, and inference. Within this framework, we implement a broad suite of attacks, including all existing attacks targeting VLAs and multiple adapted attacks originally developed for vision-language models, and evaluate them in both simulation and real-world settings. Our analysis of existing attacks reveals a critical gap: current methods tend to induce untargeted failures or static action states, leaving targeted attacks that drive VLAs to perform precise long-horizon action sequences largely unexplored. To fill this gap, we introduce BackdoorVLA, a targeted backdoor attack that compels a VLA to execute an attacker-specified long-horizon action sequence whenever a trigger is present. We evaluate BackdoorVLA in both simulated benchmarks and real-world robotic settings, achieving an average targeted success rate of 58.4% and reaching 100% on selected tasks. Our work provides a standardized framework for evaluating VLA vulnerabilities and demonstrates the potential for precise adversarial manipulation, motivating further research on securing VLA-based embodied systems.

Hanxun Huang, Sarah Erfani, Yige Li et al.

Contrastive language-image pretraining (CLIP) has been found to be vulnerable to poisoning backdoor attacks where the adversary can achieve an almost perfect attack success rate on CLIP models by poisoning only 0.01\% of the training dataset. This raises security concerns on the current practice of pretraining large-scale models on unscrutinized web data using CLIP. In this work, we analyze the representations of backdoor-poisoned samples learned by CLIP models and find that they exhibit unique characteristics in their local subspace, i.e., their local neighborhoods are far more sparse than that of clean samples. Based on this finding, we conduct a systematic study on detecting CLIP backdoor attacks and show that these attacks can be easily and efficiently detected by traditional density ratio-based local outlier detectors, whereas existing backdoor sample detection methods fail. Our experiments also reveal that an unintentional backdoor already exists in the original CC3M dataset and has been trained into a popular open-source model released by OpenCLIP. Based on our detector, one can clean up a million-scale web dataset (e.g., CC3M) efficiently within 15 minutes using 4 Nvidia A100 GPUs. The code is publicly available in our \href{https://github.com/HanxunH/Detect-CLIP-Backdoor-Samples}{GitHub repository}.

Hanxun Huang, Sarah Erfani, Yige Li et al.

As Contrastive Language-Image Pre-training (CLIP) models are increasingly adopted for diverse downstream tasks and integrated into large vision-language models (VLMs), their susceptibility to adversarial perturbations has emerged as a critical concern. In this work, we introduce \textbf{X-Transfer}, a novel attack method that exposes a universal adversarial vulnerability in CLIP. X-Transfer generates a Universal Adversarial Perturbation (UAP) capable of deceiving various CLIP encoders and downstream VLMs across different samples, tasks, and domains. We refer to this property as \textbf{super transferability}--a single perturbation achieving cross-data, cross-domain, cross-model, and cross-task adversarial transferability simultaneously. This is achieved through \textbf{surrogate scaling}, a key innovation of our approach. Unlike existing methods that rely on fixed surrogate models, which are computationally intensive to scale, X-Transfer employs an efficient surrogate scaling strategy that dynamically selects a small subset of suitable surrogates from a large search space. Extensive evaluations demonstrate that X-Transfer significantly outperforms previous state-of-the-art UAP methods, establishing a new benchmark for adversarial transferability across CLIP models. The code is publicly available in our \href{https://github.com/HanxunH/XTransferBench}{GitHub repository}.

Xiang Zheng, Longxiang Wang, Yi Liu et al.

Auditing Large Language Models (LLMs) is a crucial and challenging task. In this study, we focus on auditing black-box LLMs without access to their parameters, only to the provided service. We treat this type of auditing as a black-box optimization problem where the goal is to automatically uncover input-output pairs of the target LLMs that exhibit illegal, immoral, or unsafe behaviors. For instance, we may seek a non-toxic input that the target LLM responds to with a toxic output or an input that induces the hallucinative response from the target LLM containing politically sensitive individuals. This black-box optimization is challenging due to the scarcity of feasible points, the discrete nature of the prompt space, and the large search space. To address these challenges, we propose Curiosity-Driven Auditing for Large Language Models (CALM), which uses intrinsically motivated reinforcement learning to finetune an LLM as the auditor agent to uncover potential harmful and biased input-output pairs of the target LLM. CALM successfully identifies derogatory completions involving celebrities and uncovers inputs that elicit specific names under the black-box setting. This work offers a promising direction for auditing black-box LLMs. Our code is available at https://github.com/x-zheng16/CALM.git.

Yixu Wang, Jiaxin Song, Yifeng Gao et al.

As Video Large Multimodal Models (VLMMs) rapidly advance, their inherent complexity introduces significant safety challenges, particularly the issue of mismatched generalization where static safety alignments fail to transfer to dynamic video contexts. We introduce SafeVid, a framework designed to instill video-specific safety principles in VLMMs. SafeVid uniquely transfers robust textual safety alignment capabilities to the video domain by employing detailed textual video descriptions as an interpretive bridge, facilitating LLM-based rule-driven safety reasoning. This is achieved through a closed-loop system comprising: 1) generation of SafeVid-350K, a novel 350,000-pair video-specific safety preference dataset; 2) targeted alignment of VLMMs using Direct Preference Optimization (DPO); and 3) comprehensive evaluation via our new SafeVidBench benchmark. Alignment with SafeVid-350K significantly enhances VLMM safety, with models like LLaVA-NeXT-Video demonstrating substantial improvements (e.g., up to 42.39%) on SafeVidBench. SafeVid provides critical resources and a structured approach, demonstrating that leveraging textual descriptions as a conduit for safety reasoning markedly improves the safety alignment of VLMMs. We have made SafeVid-350K dataset (https://huggingface.co/datasets/yxwang/SafeVid-350K) publicly available.

Yisen Wang, Yichuan Mo, Dongxian Wu et al.

Skip connection is an essential ingredient for modern deep models to be deeper and more powerful. Despite their huge success in normal scenarios (state-of-the-art classification performance on natural examples), we investigate and identify an interesting property of skip connections under adversarial scenarios, namely, the use of skip connections allows easier generation of highly transferable adversarial examples. Specifically, in ResNet-like models (with skip connections), we find that using more gradients from the skip connections rather than the residual modules according to a decay factor during backpropagation allows one to craft adversarial examples with high transferability. The above method is termed as Skip Gradient Method (SGM). Although starting from ResNet-like models in vision domains, we further extend SGM to more advanced architectures, including Vision Transformers (ViTs) and models with length-varying paths and other domains, i.e. natural language processing. We conduct comprehensive transfer attacks against various models including ResNets, Transformers, Inceptions, Neural Architecture Search, and Large Language Models (LLMs). We show that employing SGM can greatly improve the transferability of crafted attacks in almost all cases. Furthermore, considering the big complexity for practical use, we further demonstrate that SGM can even improve the transferability on ensembles of models or targeted attacks and the stealthiness against current defenses. At last, we provide theoretical explanations and empirical insights on how SGM works. Our findings not only motivate new adversarial research into the architectural characteristics of models but also open up further challenges for secure model architecture design. Our code is available at https://github.com/mo666666/SGM.

Yige Li, Hanxun Huang, Jiaming Zhang et al.

Backdoor attacks covertly implant triggers into deep neural networks (DNNs) by poisoning a small portion of the training data with pre-designed backdoor triggers. This vulnerability is exacerbated in the era of large models, where extensive (pre-)training on web-crawled datasets is susceptible to compromise. In this paper, we introduce a novel two-step defense framework named Expose Before You Defend (EBYD). EBYD unifies existing backdoor defense methods into a comprehensive defense system with enhanced performance. Specifically, EBYD first exposes the backdoor functionality in the backdoored model through a model preprocessing step called backdoor exposure, and then applies detection and removal methods to the exposed model to identify and eliminate the backdoor features. In the first step of backdoor exposure, we propose a novel technique called Clean Unlearning (CUL), which proactively unlearns clean features from the backdoored model to reveal the hidden backdoor features. We also explore various model editing/modification techniques for backdoor exposure, including fine-tuning, model sparsification, and weight perturbation. Using EBYD, we conduct extensive experiments on 10 image attacks and 6 text attacks across 2 vision datasets (CIFAR-10 and an ImageNet subset) and 4 language datasets (SST-2, IMDB, Twitter, and AG's News). The results demonstrate the importance of backdoor exposure for backdoor defense, showing that the exposed models can significantly benefit a range of downstream defense tasks, including backdoor label detection, backdoor trigger recovery, backdoor model detection, and backdoor removal. We hope our work could inspire more research in developing advanced defense frameworks with exposed models. Our code is available at: https://github.com/bboylyg/Expose-Before-You-Defend.

Yunhao Chen, Xin Wang, Juncheng Li et al.

Automated red teaming frameworks for Large Language Models (LLMs) have become increasingly sophisticated, yet they share a fundamental limitation: their jailbreak logic is confined to selecting, combining, or refining pre-existing attack strategies. This binds their creativity and leaves them unable to autonomously invent entirely new attack mechanisms. To overcome this gap, we introduce \textbf{EvoSynth}, an autonomous framework that shifts the paradigm from attack planning to the evolutionary synthesis of jailbreak methods. Instead of refining prompts, EvoSynth employs a multi-agent system to autonomously engineer, evolve, and execute novel, code-based attack algorithms. Crucially, it features a code-level self-correction loop, allowing it to iteratively rewrite its own attack logic in response to failure. Through extensive experiments, we demonstrate that EvoSynth not only establishes a new state-of-the-art by achieving an 85.5\% Attack Success Rate (ASR) against highly robust models like Claude-Sonnet-4.5, but also generates attacks that are significantly more diverse than those from existing methods. We release our framework to facilitate future research in this new direction of evolutionary synthesis of jailbreak methods. Code is available at: https://github.com/dongdongunique/EvoSynth.

Kuofeng Gao, Yiming Li, Chao Du et al.

Jailbreaking attacks on the vision modality typically rely on imperceptible adversarial perturbations, whereas attacks on the textual modality are generally assumed to require visible modifications (e.g., non-semantic suffixes). In this paper, we introduce imperceptible jailbreaks that exploit a class of Unicode characters called variation selectors. By appending invisible variation selectors to malicious questions, the jailbreak prompts appear visually identical to original malicious questions on screen, while their tokenization is "secretly" altered. We propose a chain-of-search pipeline to generate such adversarial suffixes to induce harmful responses. Our experiments show that our imperceptible jailbreaks achieve high attack success rates against four aligned LLMs and generalize to prompt injection attacks, all without producing any visible modifications in the written prompt. Our code is available at https://github.com/sail-sg/imperceptible-jailbreaks.

Ruofan Wang, Xin Wang, Yang Yao et al.

Fine-tuning open-source Vision-Language Models (VLMs) creates a critical yet underexplored attack surface: vulnerabilities in the base VLM could be retained in fine-tuned variants, rendering them susceptible to transferable jailbreak attacks. To demonstrate this risk, we introduce the Simulated Ensemble Attack (SEA), a novel grey-box jailbreak method in which the adversary has full access to the base VLM but no knowledge of the fine-tuned target's weights or training configuration. To improve jailbreak transferability across fine-tuned VLMs, SEA combines two key techniques: Fine-tuning Trajectory Simulation (FTS) and Targeted Prompt Guidance (TPG). FTS generates transferable adversarial images by simulating the vision encoder's parameter shifts, while TPG is a textual strategy that steers the language decoder toward adversarially optimized outputs. Experiments on the Qwen2-VL family (2B and 7B) demonstrate that SEA achieves high transfer attack success rates exceeding 86.5% and toxicity rates near 49.5% across diverse fine-tuned variants, even those specifically fine-tuned to improve safety behaviors. Notably, while direct PGD-based image jailbreaks rarely transfer across fine-tuned VLMs, SEA reliably exploits inherited vulnerabilities from the base model, significantly enhancing transferability. These findings highlight an urgent need to safeguard fine-tuned proprietary VLMs against transferable vulnerabilities inherited from open-source foundations, motivating the development of holistic defenses across the entire model lifecycle.

Xiang Zheng, Xingjun Ma, Wei-Bin Lee et al.

Red teaming has proven to be an effective method for identifying and mitigating vulnerabilities in Large Language Models (LLMs). Reinforcement Fine-Tuning (RFT) has emerged as a promising strategy among existing red teaming techniques. However, a lack of a unified benchmark hinders current RFT-based red teaming methods. Implementation details, especially in Proximal Policy Optimization (PPO)-based RFT, significantly affect outcome stability and reproducibility. To address this issue, we introduce RedRFT, a lightweight benchmark designed to simplify and standardize the implementation and evaluation of RFT-based red teaming. RedRFT combines the design strengths of both single-file CleanRL and highly modularized Tianshou, offering high-quality single-file red teaming implementations and modular PPO core components, such as the General Advantage Estimator. It supports a variety of token and sentence diversity metrics, featuring modularized intrinsic reward computation that facilitates plug-and-play experimentation. To clarify their influence on RFT performance, we conducted an extensive ablation study on key components, including Low-Rank Adaptation (LoRA), Kullback-Leibler (KL) divergence, and Lagrange Multiplier. We hope this work contributes to 1) gaining a comprehensive understanding of the implementation nuances of RFT-based red teaming algorithms, and 2) enabling rapid prototyping of innovative features for RFT-based red teaming. Code for the benchmark can be accessed at https://github.com/x-zheng16/RedRFT.git.

Xincheng Shuai, Henghui Ding, Xingjun Ma et al.

Image editing aims to edit the given synthetic or real image to meet the specific requirements from users. It is widely studied in recent years as a promising and challenging field of Artificial Intelligence Generative Content (AIGC). Recent significant advancement in this field is based on the development of text-to-image (T2I) diffusion models, which generate images according to text prompts. These models demonstrate remarkable generative capabilities and have become widely used tools for image editing. T2I-based image editing methods significantly enhance editing performance and offer a user-friendly interface for modifying content guided by multimodal inputs. In this survey, we provide a comprehensive review of multimodal-guided image editing techniques that leverage T2I diffusion models. First, we define the scope of image editing from a holistic perspective and detail various control signals and editing scenarios. We then propose a unified framework to formalize the editing process, categorizing it into two primary algorithm families. This framework offers a design space for users to achieve specific goals. Subsequently, we present an in-depth analysis of each component within this framework, examining the characteristics and applicable scenarios of different combinations. Given that training-based methods learn to directly map the source image to target one under user guidance, we discuss them separately, and introduce injection schemes of source image in different scenarios. Additionally, we review the application of 2D techniques to video editing, highlighting solutions for inter-frame inconsistency. Finally, we discuss open challenges in the field and suggest potential future research directions. We keep tracing related works at https://github.com/xinchengshuai/Awesome-Image-Editing.

Yige Li, Xixiang Lyu, Xingjun Ma et al.

Deep neural networks (DNNs) have been found to be vulnerable to backdoor attacks, raising security concerns about their deployment in mission-critical applications. While existing defense methods have demonstrated promising results, it is still not clear how to effectively remove backdoor-associated neurons in backdoored DNNs. In this paper, we propose a novel defense called \emph{Reconstructive Neuron Pruning} (RNP) to expose and prune backdoor neurons via an unlearning and then recovering process. Specifically, RNP first unlearns the neurons by maximizing the model's error on a small subset of clean samples and then recovers the neurons by minimizing the model's error on the same data. In RNP, unlearning is operated at the neuron level while recovering is operated at the filter level, forming an asymmetric reconstructive learning procedure. We show that such an asymmetric process on only a few clean samples can effectively expose and prune the backdoor neurons implanted by a wide range of attacks, achieving a new state-of-the-art defense performance. Moreover, the unlearned model at the intermediate step of our RNP can be directly used to improve other backdoor defense tasks including backdoor removal, trigger recovery, backdoor label detection, and backdoor sample detection. Code is available at \url{https://github.com/bboylyg/RNP}.

Xiang Zheng, Xingjun Ma, Shengjie Wang et al.

Reinforcement learning agents are susceptible to evasion attacks during deployment. In single-agent environments, these attacks can occur through imperceptible perturbations injected into the inputs of the victim policy network. In multi-agent environments, an attacker can manipulate an adversarial opponent to influence the victim policy's observations indirectly. While adversarial policies offer a promising technique to craft such attacks, current methods are either sample-inefficient due to poor exploration strategies or require extra surrogate model training under the black-box assumption. To address these challenges, in this paper, we propose Intrinsically Motivated Adversarial Policy (IMAP) for efficient black-box adversarial policy learning in both single- and multi-agent environments. We formulate four types of adversarial intrinsic regularizers -- maximizing the adversarial state coverage, policy coverage, risk, or divergence -- to discover potential vulnerabilities of the victim policy in a principled way. We also present a novel bias-reduction method to balance the extrinsic objective and the adversarial intrinsic regularizers adaptively. Our experiments validate the effectiveness of the four types of adversarial intrinsic regularizers and the bias-reduction method in enhancing black-box adversarial policy learning across a variety of environments. Our IMAP successfully evades two types of defense methods, adversarial training and robust regularizer, decreasing the performance of the state-of-the-art robust WocaR-PPO agents by 34\%-54\% across four single-agent tasks. IMAP also achieves a state-of-the-art attacking success rate of 83.91\% in the multi-agent game YouShallNotPass. Our code is available at \url{https://github.com/x-zheng16/IMAP}.

Yige Li, Xixiang Lyu, Nodens Koren et al.

Backdoor attack has emerged as a major security threat to deep neural networks (DNNs). While existing defense methods have demonstrated promising results on detecting or erasing backdoors, it is still not clear whether robust training methods can be devised to prevent the backdoor triggers being injected into the trained model in the first place. In this paper, we introduce the concept of \emph{anti-backdoor learning}, aiming to train \emph{clean} models given backdoor-poisoned data. We frame the overall learning process as a dual-task of learning the \emph{clean} and the \emph{backdoor} portions of data. From this view, we identify two inherent characteristics of backdoor attacks as their weaknesses: 1) the models learn backdoored data much faster than learning with clean data, and the stronger the attack the faster the model converges on backdoored data; 2) the backdoor task is tied to a specific class (the backdoor target class). Based on these two weaknesses, we propose a general learning scheme, Anti-Backdoor Learning (ABL), to automatically prevent backdoor attacks during training. ABL introduces a two-stage \emph{gradient ascent} mechanism for standard training to 1) help isolate backdoor examples at an early training stage, and 2) break the correlation between backdoor examples and the target class at a later training stage. Through extensive experiments on multiple benchmark datasets against 10 state-of-the-art attacks, we empirically show that ABL-trained models on backdoor-poisoned data achieve the same performance as they were trained on purely clean data. Code is available at \url{https://github.com/bboylyg/ABL}.

Hanxun Huang, Yisen Wang, Sarah Monazam Erfani et al.

Deep neural networks (DNNs) are known to be vulnerable to adversarial attacks. A range of defense methods have been proposed to train adversarially robust DNNs, among which adversarial training has demonstrated promising results. However, despite preliminary understandings developed for adversarial training, it is still not clear, from the architectural perspective, what configurations can lead to more robust DNNs. In this paper, we address this gap via a comprehensive investigation on the impact of network width and depth on the robustness of adversarially trained DNNs. Specifically, we make the following key observations: 1) more parameters (higher model capacity) does not necessarily help adversarial robustness; 2) reducing capacity at the last stage (the last group of blocks) of the network can actually improve adversarial robustness; and 3) under the same parameter budget, there exists an optimal architectural configuration for adversarial robustness. We also provide a theoretical analysis explaning why such network configuration can help robustness. These architectural insights can help design adversarially robust DNNs. Code is available at \url{https://github.com/HanxunH/RobustWRN}.

Yige Li, Xixiang Lyu, Nodens Koren et al.

Deep neural networks (DNNs) are known vulnerable to backdoor attacks, a training time attack that injects a trigger pattern into a small proportion of training data so as to control the model's prediction at the test time. Backdoor attacks are notably dangerous since they do not affect the model's performance on clean examples, yet can fool the model to make incorrect prediction whenever the trigger pattern appears during testing. In this paper, we propose a novel defense framework Neural Attention Distillation (NAD) to erase backdoor triggers from backdoored DNNs. NAD utilizes a teacher network to guide the finetuning of the backdoored student network on a small clean subset of data such that the intermediate-layer attention of the student network aligns with that of the teacher network. The teacher network can be obtained by an independent finetuning process on the same clean subset. We empirically show, against 6 state-of-the-art backdoor attacks, NAD can effectively erase the backdoor triggers using only 5\% clean training data without causing obvious performance degradation on clean examples. Code is available in https://github.com/bboylyg/NAD.

Bojia Zi, Minghao Chang, Jingjing Chen et al.

In recent years, the abuse of a face swap technique called deepfake has raised enormous public concerns. So far, a large number of deepfake videos (known as "deepfakes") have been crafted and uploaded to the internet, calling for effective countermeasures. One promising countermeasure against deepfakes is deepfake detection. Several deepfake datasets have been released to support the training and testing of deepfake detectors, such as DeepfakeDetection and FaceForensics++. While this has greatly advanced deepfake detection, most of the real videos in these datasets are filmed with a few volunteer actors in limited scenes, and the fake videos are crafted by researchers using a few popular deepfake softwares. Detectors developed on these datasets may become less effective against real-world deepfakes on the internet. To better support detection against real-world deepfakes, in this paper, we introduce a new dataset WildDeepfake which consists of 7,314 face sequences extracted from 707 deepfake videos collected completely from the internet. WildDeepfake is a small dataset that can be used, in addition to existing datasets, to develop and test the effectiveness of deepfake detectors against real-world deepfakes. We conduct a systematic evaluation of a set of baseline detection networks on both existing and our WildDeepfake datasets, and show that WildDeepfake is indeed a more challenging dataset, where the detection performance can decrease drastically. We also propose two (eg. 2D and 3D) Attention-based Deepfake Detection Networks (ADDNets) to leverage the attention masks on real/fake faces for improved detection. We empirically verify the effectiveness of ADDNets on both existing datasets and WildDeepfake. The dataset is available at: https://github.com/OpenTAI/wild-deepfake.

Xingjun Ma, Linxi Jiang, Hanxun Huang et al.

Evaluating the robustness of a defense model is a challenging task in adversarial robustness research. Obfuscated gradients have previously been found to exist in many defense methods and cause a false signal of robustness. In this paper, we identify a more subtle situation called Imbalanced Gradients that can also cause overestimated adversarial robustness. The phenomenon of imbalanced gradients occurs when the gradient of one term of the margin loss dominates and pushes the attack towards to a suboptimal direction. To exploit imbalanced gradients, we formulate a Margin Decomposition (MD) attack that decomposes a margin loss into individual terms and then explores the attackability of these terms separately via a two-stage process. We also propose a multi-targeted and ensemble version of our MD attack. By investigating 24 defense models proposed since 2018, we find that 11 models are susceptible to a certain degree of imbalanced gradients and our MD attack can decrease their robustness evaluated by the best standalone baseline attack by more than 1%. We also provide an in-depth investigation on the likely causes of imbalanced gradients and effective countermeasures. Our code is available at https://github.com/HanxunH/MDAttack.

Kun Wang, Guibin Zhang, Zhenhong Zhou et al. · mit

The remarkable success of Large Language Models (LLMs) has illuminated a promising pathway toward achieving Artificial General Intelligence for both academic and industrial communities, owing to their unprecedented performance across various applications. As LLMs continue to gain prominence in both research and commercial domains, their security and safety implications have become a growing concern, not only for researchers and corporations but also for every nation. Currently, existing surveys on LLM safety primarily focus on specific stages of the LLM lifecycle, e.g., deployment phase or fine-tuning phase, lacking a comprehensive understanding of the entire "lifechain" of LLMs. To address this gap, this paper introduces, for the first time, the concept of "full-stack" safety to systematically consider safety issues throughout the entire process of LLM training, deployment, and eventual commercialization. Compared to the off-the-shelf LLM safety surveys, our work demonstrates several distinctive advantages: (I) Comprehensive Perspective. We define the complete LLM lifecycle as encompassing data preparation, pre-training, post-training, deployment and final commercialization. To our knowledge, this represents the first safety survey to encompass the entire lifecycle of LLMs. (II) Extensive Literature Support. Our research is grounded in an exhaustive review of over 800+ papers, ensuring comprehensive coverage and systematic organization of security issues within a more holistic understanding. (III) Unique Insights. Through systematic literature analysis, we have developed reliable roadmaps and perspectives for each chapter. Our work identifies promising research directions, including safety in data generation, alignment techniques, model editing, and LLM-based agent systems. These insights provide valuable guidance for researchers pursuing future work in this field.

Yixin Cao, Shibo Hong, Xinze Li et al.

Large Language Models (LLMs) are advancing at an amazing speed and have become indispensable across academia, industry, and daily applications. To keep pace with the status quo, this survey probes the core challenges that the rise of LLMs poses for evaluation. We identify and analyze two pivotal transitions: (i) from task-specific to capability-based evaluation, which reorganizes benchmarks around core competencies such as knowledge, reasoning, instruction following, multi-modal understanding, and safety; and (ii) from manual to automated evaluation, encompassing dynamic dataset curation and "LLM-as-a-judge" scoring. Yet, even with these transitions, a crucial obstacle persists: the evaluation generalization issue. Bounded test sets cannot scale alongside models whose abilities grow seemingly without limit. We will dissect this issue, along with the core challenges of the above two transitions, from the perspectives of methods, datasets, evaluators, and metrics. Due to the fast evolving of this field, we will maintain a living GitHub repository (links are in each section) to crowd-source updates and corrections, and warmly invite contributors and collaborators.

Pagnarasmey Pit, Xingjun Ma, Mike Conway et al.

Large Language Models (LLMs) have gained significant popularity for their application in various everyday tasks such as text generation, summarization, and information retrieval. As the widespread adoption of LLMs continues to surge, it becomes increasingly crucial to ensure that these models yield responses that are politically impartial, with the aim of preventing information bubbles, upholding fairness in representation, and mitigating confirmation bias. In this paper, we propose a quantitative framework and pipeline designed to systematically investigate the political orientation of LLMs. Our investigation delves into the political alignment of LLMs across a spectrum of eight polarizing topics, spanning from abortion to LGBTQ issues. Across topics, the results indicate that LLMs exhibit a tendency to provide responses that closely align with liberal or left-leaning perspectives rather than conservative or right-leaning ones when user queries include details pertaining to occupation, race, or political affiliation. The findings presented in this study not only reaffirm earlier observations regarding the left-leaning characteristics of LLMs but also surface particular attributes, such as occupation, that are particularly susceptible to such inclinations even when directly steered towards conservatism. As a recommendation to avoid these models providing politicised responses, users should be mindful when crafting queries, and exercise caution in selecting neutral prompt language.

Xin Wang, Kai Chen, Jiaming Zhang et al.

Large pre-trained Vision-Language Models (VLMs) such as CLIP have demonstrated excellent zero-shot generalizability across various downstream tasks. However, recent studies have shown that the inference performance of CLIP can be greatly degraded by small adversarial perturbations, especially its visual modality, posing significant safety threats. To mitigate this vulnerability, in this paper, we propose a novel defense method called Test-Time Adversarial Prompt Tuning (TAPT) to enhance the inference robustness of CLIP against visual adversarial attacks. TAPT is a test-time defense method that learns defensive bimodal (textual and visual) prompts to robustify the inference process of CLIP. Specifically, it is an unsupervised method that optimizes the defensive prompts for each test sample by minimizing a multi-view entropy and aligning adversarial-clean distributions. We evaluate the effectiveness of TAPT on 11 benchmark datasets, including ImageNet and 10 other zero-shot datasets, demonstrating that it enhances the zero-shot adversarial robustness of the original CLIP by at least 48.9% against AutoAttack (AA), while largely maintaining performance on clean examples. Moreover, TAPT outperforms existing adversarial prompt tuning methods across various backbones, achieving an average robustness improvement of at least 36.6%.

Ye Sun, Hao Zhang, Tiehua Zhang et al.

Image segmentation is a crucial vision task that groups pixels within an image into semantically meaningful segments, which is pivotal in obtaining a fine-grained understanding of real-world scenes. However, an increasing privacy concern exists regarding training large-scale image segmentation models on unauthorized private data. In this work, we exploit the concept of unlearnable examples to make images unusable to model training by generating and adding unlearnable noise into the original images. Particularly, we propose a novel Unlearnable Segmentation (UnSeg) framework to train a universal unlearnable noise generator that is capable of transforming any downstream images into their unlearnable version. The unlearnable noise generator is finetuned from the Segment Anything Model (SAM) via bilevel optimization on an interactive segmentation dataset towards minimizing the training error of a surrogate model that shares the same architecture with SAM but is trained from scratch. We empirically verify the effectiveness of UnSeg across 6 mainstream image segmentation tasks, 10 widely used datasets, and 7 different network architectures, and show that the unlearnable images can reduce the segmentation performance by a large margin. Our work provides useful insights into how to leverage foundation models in a data-efficient and computationally affordable manner to protect images against image segmentation models.

Ruofan Wang, Juncheng Li, Yixu Wang et al.

As large Vision-Language Models (VLMs) gain prominence, ensuring their safe deployment has become critical. Recent studies have explored VLM robustness against jailbreak attacks-techniques that exploit model vulnerabilities to elicit harmful outputs. However, the limited availability of diverse multimodal data has constrained current approaches to rely heavily on adversarial or manually crafted images derived from harmful text datasets, which often lack effectiveness and diversity across different contexts. In this paper, we propose IDEATOR, a novel jailbreak method that autonomously generates malicious image-text pairs for black-box jailbreak attacks. IDEATOR is grounded in the insight that VLMs themselves could serve as powerful red team models for generating multimodal jailbreak prompts. Specifically, IDEATOR leverages a VLM to create targeted jailbreak texts and pairs them with jailbreak images generated by a state-of-the-art diffusion model. Extensive experiments demonstrate IDEATOR's high effectiveness and transferability, achieving a 94% attack success rate (ASR) in jailbreaking MiniGPT-4 with an average of only 5.34 queries, and high ASRs of 82%, 88%, and 75% when transferred to LLaVA, InstructBLIP, and Chameleon, respectively. Building on IDEATOR's strong transferability and automated process, we introduce the VLJailbreakBench, a safety benchmark comprising 3,654 multimodal jailbreak samples. Our benchmark results on 11 recently released VLMs reveal significant gaps in safety alignment. For instance, our challenge set achieves ASRs of 46.31% on GPT-4o and 19.65% on Claude-3.5-Sonnet, underscoring the urgent need for stronger defenses. VLJailbreakBench is publicly available at https://roywang021.github.io/VLJailbreakBench.

Lin Luo, Xin Wang, Bojia Zi et al.

Large pre-trained Vision-Language Models (VLMs) such as Contrastive Language-Image Pre-training (CLIP) have been shown to be susceptible to adversarial attacks, raising concerns about their deployment in safety-critical applications like autonomous driving and medical diagnosis. One promising approach for robustifying pre-trained VLMs is Adversarial Prompt Tuning (APT), which applies adversarial training during the process of prompt tuning. However, existing APT methods are mostly single-modal methods that design prompt(s) for only the visual or textual modality, limiting their effectiveness in either robustness or clean accuracy. In this work, we propose Adversarial Prompt Distillation (APD), a bimodal knowledge distillation framework that enhances APT by integrating it with multi-modal knowledge transfer. APD optimizes prompts for both visual and textual modalities while distilling knowledge from a clean pre-trained teacher CLIP model. Extensive experiments on multiple benchmark datasets demonstrate the superiority of our APD method over the current state-of-the-art APT methods in terms of both adversarial robustness and clean accuracy. The effectiveness of APD also validates the possibility of using a non-robust teacher to improve the generalization and robustness of fine-tuned VLMs.