Adversarial Transferability

The work presents an approach for addressing the challenge of robustness in Large Language Models (LLMs) to alterations and potential errors caused by semantically similar but textually different prompts. Recent works have shown that these kinds of prompt variations can significantly impact the performance of LLMs on tasks. The central question is: can LLMs' robustness to semantically-neutral prompt alterations be acquired without expensive retraining of the entire model? We address this question both theoretically and through experiments. Our theoretical analysis reveals a crucial factor impacting model robustness - a systematic expected shift or perturbation-induced bias in neural network module outputs. Motivated by this analysis, we show that robustness can be achieved via a simple fine-tuning process: debiasing for robustness. We identify conditions when debiasing helps and when it does not, and demonstrate, through both theory and extensive experiments, that debiasing for robustness may indeed be a quick and efficient tool to enhance robustness and provide certification against random prompt perturbations.

While Vision-Language-Action (VLA) models have emerged as powerful generalist policies, their severe vulnerability to adversarial patches significantly hinders their deployment in safety-critical domains. Moreover, existing patch attacks primarily focus on white-box settings, heavily overfitting to the specific action output space of the target model, which results in poor cross-architecture transferability. To overcome this limitation, we propose VLA-Hijack, a unified adversarial framework that breaks the transferability bottleneck by exploiting a fundamental vulnerability identified in this work: before planning any motion, a VLA model must first use visual information to locate its own robotic arm within the environment. Targeting this shared visual self-localization process, our approach concurrently optimizes Attention-Guided Proprioceptive Suppression to inhibit the real robotic arm's features, and Multimodal Proprioceptive Injection to establish the patch as a surrogate "phantom embodiment". By alternating between semantic concept anchoring and visual prototype projection, VLA-Hijack effectively severs the semantic relationship between the agent's true embodiment and its control policy. Extensive experiments across diverse architectures (OpenVLA, UniVLA, and CronusVLA) demonstrate that VLA-Hijack achieves superior optimization efficiency in white-box settings and sets a new SOTA for cross-architecture and cross-domain black-box transferability.

Ultra-low-field (ULF) MRI offers portable and low-cost imaging but suffers from poor image quality. To address this, we present our submission to the 2025 ULF Enhancement Challenge (ULF-EnC), where the goal is to synthesise high-field-like MRIs from 64 mT scans. Our pipeline enhances ULF MRI through a combination of anatomical conditioning and model ensembling. We first generate tissue segmentation priors using a Swin UNETR trained solely on challenge-provided data. These priors condition two independent enhancement networks - a CycleGAN and a transformer-based residual enhancement model (T-REX) - each trained to synthesise 3 T-like MRIs. Outputs from both models are combined using a weighted average. Our approach produces enhanced MRIs that were comparable to high-field scans both quantitatively and qualitatively.

This paper describes our approach for the 8th UG2+ Workshop (CVPR 2026) Track~2, which targets semantic segmentation of outdoor scenes degraded by five weather conditions: blur, darkness, snow, haze, and glare. A central challenge we observe is a severe generalization gap -- models that perform well on the validation set often collapse on the test set. For instance, SegFormer-B5 drops 16.1 mIoU points from validation to test, suggesting that model capacity alone is insufficient for robustness. We investigate whether a carefully designed training recipe, rather than architectural complexity, can address this gap. Starting from a pre-trained SegMAN-S backbone, we systematically study the effects of domain-adaptive fine-tuning, multi-source data mixing, scene-balanced sampling, and synthetic degradation augmentation. Our final system achieves 59.9% mIoU on the official test set while maintaining a validation-test gap of only 6.5 points -- less than half that of larger models. We analyze negative results from architectural modifications, loss function variants, and model scaling to provide practical insights for weather-robust segmentation under limited data.

Vision Language Models adapt well to downstream tasks but are highly vulnerable to adversarial perturbations that disrupt cross-modal semantic alignment. Existing defenses are largely unidirectional or structural, failing to exploit bidirectional cross-modal complementarity and instance-wise adaptive protection. To overcome the limitations of unidirectional and static defenses in adversarial settings, we propose Closed-Loop Bidirectional Prompting, casting robust adaptation as cross-modal agreement recovery via a dynamic feedback loop on frozen encoders. A Semantic Anchor is introduced as a stable prior to constrain cyclic updates and mitigate perturbation-induced feature corruption. Through anchor-based bootstrapping, textual semantics denoise visual representations, while the refined visuals enable instance-adaptive prompt updating, yielding a rectified and robust consensus. Extensive evaluations across 11 datasets validate state-of-the-art robustness and strong base-to-new generalization, while maintaining a favorable trade-off between computational cost and accuracy.

Reinforcement learning with verifiable rewards (RLVR) has driven breakthroughs in domains such as math, tool-use, and software engineering, yet its extension to computer-use agents (CUAs) has been bottlenecked by the scarcity of scalable training data with deterministic rewards. Constructing such data for CUAs requires consistent task instruction, executable environment, and verifiable reward. However, hand-curated benchmarks achieve high reward fidelity but cover few applications and LLM-as-judge-based datasets scale broadly but lack reliable verification. We present CUA-Gym, a scalable pipeline that co-generates task instructions, environment states, and reward functions. Concretely, a Generator agent constructs the initial and golden environment states, and a separate Discriminator agent writes the reward function from the task specification. An orchestrator agent drives the two through iterative rounds upon execution. Generated tuples then pass a final filter combining LLM majority voting and agent rollouts, ensuring quality beyond the per-task adversarial loop. To address the scarcity of training environments, we further synthesize CUA-Gym-Hub, a broad suite of high-fidelity mock web applications grounded in real-world software-use distributions, expanding the scale of CUA RLVR data by magnitude. Using this pipeline, we construct CUA-Gym, a dataset of 32,112 verified RLVR training tuples grounded in 110 environments. Trained with GSPO on CUA-Gym, our CUA-Gym-A3B and CUA-Gym-A17B achieve 62.1% and 72.6% on OSWorld-Verified, outperforming prior open-source CUAs at comparable scales, with performance scaling smoothly in both data volume and environment diversity. The same checkpoints also improve on the held-out WebArena benchmark, indicating transfer beyond the training environments. We will open-source the full synthesis pipeline, dataset, CUA-Gym-Hub environments, and models.

Deep learning models are vulnerable to adversarial perturbations, raising important concerns for safety-critical deployment. Empirical defenses can achieve strong robustness in practice, but lack formal guarantees, motivating the need for certifiably robust classifiers. While certified methods provide formal guarantees, they often yield overly conservative bounds due to their inability to exploit structure in complex data distributions. In this work, we propose a framework for designing certifiably robust classifiers that leverages latent structure in data representations. We first analyze the Gaussian mixture setting, deriving necessary and sufficient conditions for the existence of robust classifiers and constructing a classifier with a closed-form robustness certificate and generalization guarantees. Our main contribution is to show that exact structure is not required: we prove that if a pretrained encoder maps inputs to a latent distribution that is $\varepsilon$-close (in KL divergence) to a Gaussian mixture, then certified accuracy degrades gracefully, with an explicit bound relating robustness under the true and approximate distributions. This result enables the direct use of pretrained models without requiring exact distributional assumptions. Empirically, our method achieves state-of-the-art or competitive certified accuracy on CIFAR-10 and ImageNet, while maintaining strong clean performance and low computational overhead. Overall, our work establishes approximate latent structure as a practical and principled route to certifiable robustness.

Despite the strong performance of deep neural networks in modern Web and language applications, they remain vulnerable to adversarial attacks, especially transferable attacks that generate adversarial examples using surrogate models without accessing the victim model. Transferable attacks in the text domain are still under-explored, with only a few studies addressing this challenging issue, often with suboptimal results due to equal treatment of submodels or inaccurate estimation of importance scores. To address these challenges, we propose a simple yet effective paradigm for transfer-based textual adversarial attack, named SEP-Attack. Specifically, we employ the Determinantal Point Process (DPP) to generate diverse surrogate ensemble weights, representing the transferability of submodels. Using these weights, we introduce a new metric to evaluate prediction confidence scores, which in turn are used to calculate word importance scores and generate adversarial candidates. Finally, we quantify the transferability score for each candidate and select the top ones as the final transferable adversarial examples. Experiments conducted on four datasets and two real-world APIs validate the efficacy of SEP-Attack, significantly outperforming state-of-the-art baselines.

The integration of fairness and privacy in centralized data-driven applications is critical, especially as these systems increasingly influence sectors with significant societal impact. Current methods rarely address privacy, fairness, and accuracy together, which can potentially compromise ethical standards and privacy regulations. However, balancing these three objectives is quite challenging since each of objective often imposes conflicting requirements on the design and training of models, making it difficult to optimize one without compromising the others. This paper introduces a novel multitask adversarial model that treats fairness and privacy as integral objectives rather than afterthoughts, and learns a latent representation that hides sensitive attributes while preserving essential task-related information. Our approach dynamically balances fairness with accuracy and privacy through an optimized cost function with minimal performance loss even under strict conditions. Extensive testing on diverse datasets shows the ability of our model to achieve high standards of fairness and privacy without significant sacrifice to accuracy. Benchmarking against state-of-the-art privacy and fairness standards shows that our method enhances the robustness of privacy, fairness, and accuracy optimization, proving its adaptability across various datasets.

Multimodal large language models (MLLMs) need efficient mechanisms to update knowledge without degrading existing capabilities. While intrinsic multimodal knowledge editing achieves strong reliability and locality, it often exhibits limited generality, failing to propagate edits across semantically equivalent visual and linguistic variations. This issue arises from the lack of explicit semantic supervision, rigid editing scopes, and biased anchoring to individual samples in high-dimensional multimodal spaces. We address robust intrinsic multimodal knowledge editing by explicitly targeting generalization. We formalize robustness through knowledge units that group semantically equivalent multimodal inputs and define generality as consistent predictions within each unit. To expose fragile semantic regions, we introduce Latent Adversarial Robustification (LAR), which generates adversarial yet semantically coherent variants in the joint latent space. We further propose Rank-Constrained Subspace Learning (RCSL), enforcing low-rank alignment of adversarial representations at the edit layer via a singular value-based objective. Extensive analysis demonstrates the effectiveness of ASAM empirically.

Multimodal large language models (MLLMs) remain vulnerable to transfer-based targeted attacks, where perturbations optimized on open-source surrogate encoders can generalize to closed-source MLLMs. A key challenge for improving adversarial transferability is to effectively capture the intrinsic visual focus shared across different models, such that perturbations align with transferable semantic cues rather than surrogate-specific behaviors. However, existing methods suffer from spatial-domain feature redundancy and surrogate-specific gradient signals, thereby hindering cross-model transferability. In this paper, we propose FRA-Attack, which addresses both challenges from a unified frequency-domain regularization perspective. For feature alignment, a high-pass DCT objective on patch features suppresses redundant global structures and concentrates the loss on the high-frequency band that carries the MLLMs' intrinsic visual focus. For gradient optimization, we introduce Frequency-domain Gradient Regularization (FGR), a \textit{model-agnostic} low-pass regularizer that modulates the surrogate gradient using only the geometric frequency coordinate, \textit{i.e.}, no surrogate-derived statistic is involved, so that FGR is model-agnostic by construction, removing surrogate-specific high-frequency artifacts while preserving transferable low-frequency directions. Together, the two components form a unified frequency-domain treatment of transferability. Extensive experiments on $15$ flagship MLLMs across $7$ vendors show that FRA-Attack achieves superior cross-model transferability, particularly with state-of-the-art performance on GPT-5.4, Claude-Opus-4.6 and Gemini-3-flash.

Dataset distillation (DD) compresses a large training set into a small synthetic set for efficient training, but most DD methods optimize only clean accuracy and leave robustness uncontrolled. Recent robust DD methods improve robustness, yet they often suffer from a poor accuracy-robustness trade-off because they (i) treat all adversarially perturbed examples uniformly, despite robust risk being dominated by near-zero robust margins, and (ii) do not explicitly increase inter-class separation in the decision boundary where attacks concentrate. We present Contrastive Curriculum for Robust Dataset Distillation (C$^2$R), a framework that couples an attack-aware curriculum with a contrastive robustness objective. From a robust-margin perspective, we derive a perturbation score that approximates each sample's robust hinge, enabling a curriculum that prioritizes the smallest-margin adversaries that most directly drive robust error. In parallel, a class-balanced contrastive robustness loss enforces adversarial invariance while explicitly widening boundary separation across classes. Experiments on CIFAR-10/100, Tiny-ImageNet, and multiple ImageNet-1K subsets under six attacks show that C$^2$R achieves the best robust accuracy, outperforming prior robust DD by $2.8$% on average.

Cooperation is central to multi-agent reinforcement learning (MARL), yet learned coordination can be fragile when external perturbations disrupt inter-agent interactions. Prior robust MARL methods have primarily considered value-oriented attacks, leaving a gap in robustness when interaction structures themselves are corrupted. In this paper, we propose an interaction-breaking adversarial learning (IBAL) framework that takes an information-theoretic view to construct attacks that impede coordination by perturbing agents' observations and actions, and trains agents to perform reliably under such disruptions. Empirically, our approach improves robustness over existing robust MARL baselines across diverse attack settings and yields stronger performance even under agent-missing scenarios. Our code is available at https://sunwoolee0504.github.io/IBAL.

Post-hoc saliency methods are widely used to interpret deep neural networks, but their faithfulness is difficult to evaluate reliably. Existing evaluations mask features according to saliency-induced feature ordering and measure performance degradation, but this degradation can be confounded by the masking operator: zero masking may create out-of-distribution artifacts, while interpolation-based masking may preserve residual predictive information. We propose Adversarial Information Masking (AIM), a saliency-guided adversarial feature replacement framework for evaluating both saliency-map faithfulness and masking-operator reliability. AIM replaces selected features with values from an adversarial counterpart of the input and compares degradation under complementary masking orders. We assess reliability using random-attribution bias and stability of explanation-method faithfulness rankings. Experiments on image, audio, and EEG tasks suggest that AIM reduces masking-induced bias compared with zero and interpolation-based masking, while revealing modality-dependent differences between signed and unsigned attributions.

Recent advances in molecular representation integrates molecular topological and visual modalities, opening new avenues for precise Molecular Relational Learning (MRL). Existing MRL methods focus on intra-domain modeling, and their inherent domain-closed effect limits applicability to molecular science, particularly in elucidating cross-domain interaction mechanisms. Consequently, the imperative for Cross-Domain Molecular Relational Learning has become increasingly pressing. Benefiting from structure-activity analysis, we propose the Domain Adversarial Training Network with Structural-Semantic Transfer Discrepancy (DisTrans) to optimize cross-domain adaptive representation for molecular structures and visual images. 1) We employ the gradient reversal strategy based on substructure topological discrepancies between domains to learn the domain dependence of molecular structures. This strategy guides the model to adapt to the structural adjacency patterns in the target domain, generating domain-separable structural representations. 2) We apply the cross-domain representation guidance mechanism to align the functional-group semantic information between the source and target domains, learning cross-domain consistency information. The experimental results in two typical cross-domain strategies demonstrate that DisTrans outperforms 16 baseline methods, maintaining satisfactory performance even under pronounced inter-domain discrepancy.

Multi-View Clustering (MVC) has gained significant attention for its ability to leverage complementary information across diverse views. However, existing deep MVC methods often struggle with view-distribution entanglement during cross-view fusion, which hampers the quality of the shared latent space and leads to suboptimal Figures. To address this issue, we propose the Generalized Multi-view Auto-Encoder (GMAE), a framework designed to preserve cross-view complementarity through disentangled representation learning. Specifically, GMAE employs dual-path autoencoders to decouple source features into view-specific and view-common embeddings, facilitating the discovery of clearer clustering structures. We further construct cross-view adversarial discriminators to guide view-specific encoders in capturing more discriminative features. By strategically modulating mutual information, GMAE effectively aligns distributions and prevents representation collapse, ensuring the generation of robust, non-trivial embeddings. Comprehensive experiments on 13 benchmark datasets demonstrate that GMAE consistently outperforms state-of-the-art methods in both complete and incomplete MVC tasks. Our code implementation is available at the repository: https://github.com/obananas/GMAE.

Vision-language models like CLIP have demonstrated remarkable zero-shot transfer capabilities. However, their susceptibility to imperceptible adversarial perturbations remains a critical security concern. While test-time defenses offer a pragmatic solution for deployed models, existing approaches typically rely on gradient-based optimization during inference, incurring significant computational overhead. In this paper, we revisit the role of data augmentation in CLIP robustness and observe that augmentations are not equally effective: specific augmentations consistently provide robust geometric cues that align with correct class semantics in the hyperspherical feature space. Based on this, we propose Adaptive Geodesic Correction (AGC), a training-free defense mechanism that requires no parameter updates. AGC identifies a reliable augmentation as a geometric anchor and corrects the input feature towards it, utilizing an adaptive step size to balance robustness against clean accuracy preservation. AGC achieves superior performance across eight fine-grained datasets and three CLIP backbones, improving average robust accuracy by 44.4% over state-of-the-art baseline while delivering a 10$\times$ reduction in inference latency. Our findings reveal a fundamental geometric property of CLIP features, offering a highly efficient and effective paradigm for robust multimodal deployment.

AI models are increasingly deployed in cloud-native environments to support scalable and automated services. However, while platforms such as Kubernetes provide strong infrastructure orchestration, security mechanisms specifically designed to protect deployed AI models remain limited. This paper presents security measures for AI models deployed in Kubernetes clusters. The proposed architecture integrates Kubeflow-based MLOps to automatically detect adversarial attacks during the inference phase and trigger defense mechanisms that preserve the model's accuracy and reliability. Specifically, a Fast Gradient Sign Method (FGSM) attack is applied at inference time, and a Projected Gradient Descent (PGD)-based adversarial training defense is automatically deployed when a degradation in accuracy is detected. The experimental results indicate that the deployed defense robustifies the model, significantly recovering accuracy relative to the degradation caused by the attack.

Deep neural networks are highly vulnerable to adversarial examples, i.e.,small perturbations that can significantly degrade model performance. While adversarial training has become the primary defense strategy, most studies focus on balanced datasets, overlooking the challenges posed by real-world long-tail data. Motivated by the fact that perturbations in adversarial examples inherently alter the training distribution, we theoretically investigate their impact. We first revisit adversarial training for long-tail data and identify two key limitations: (i) a skewed training objective caused by class imbalance, and (ii) unstable evolution of adversarial distributions. Furthermore, we show that perturbations can simultaneously address both adversarial vulnerability and class imbalance. Based on these insights, we propose RobustLT, a plug-and-play framework that adaptively adjusts perturbations during adversarial training. Extensive experiments demonstrate that RobustLT consistently enhances adversarial robustness and class-balance on long-tailed datasets. The code is available at \href{https://github.com/zhang-lilin/RobustLT}{https://github.com/zhang-lilin/RobustLT}.

One-Shot Federated Learning, where a central server learns a global model in a single communication round, has emerged as a promising paradigm. However, under extremely non-IID settings, existing data-free methods often generate low-quality data that suffers from severe semantic misalignment with ground-truth labels. To overcome these issues, we propose a novel Federated Model Inversion and Token Relabel (FedMITR) framework, which trains the global model by fully exploiting all patches of synthetic images. Specifically, FedMITR employs sparse model inversion during data generation, selectively inverting semantic foregrounds while halting the inversion of uninformative backgrounds. To address semantically meaningless tokens that hinder ViT predictions, we implement a differentiated strategy: patches with high information density utilize generated pseudo-labels, while patches with low information density are relabeled via ensemble models for robust distillation. Theoretically, our analysis based on algorithmic stability reveals that Sparse Model Inversion eliminates gradient instability arising from background noise, while Token Relabel effectively reduces gradient variance, collectively guaranteeing a tighter generalization bound. Empirically, extensive experimental results demonstrate that FedMITR substantially outperforms existing baselines under various settings.

The new wave of adversarial attacks that utilize gradient-related vulnerabilities in neural network-based classifiers makes Network Intrusion Detection Systems more open to such threats. Although state-of-the-art adversarial training methods have shown promising results in producing more robust classifiers, their interpretability and defense ability are limited due to their lack of understanding of how adversarial attacks propagate in different layers of network classifiers. In this paper, we present an insightful approach, called LARAR (Layer-wise Adversarial Robustness using Adaptive Regularization), that incorporates additional layer-wise vulnerability analysis and adaptive weighting in conventional adversarial training methods. Additionally, we utilize 'Auxiliary Classifiers' in our approach. LARAR provides interpretable layer-wise vulnerability scores, achieves a clean accuracy of 95.01%, and provides better robustness against adversarial attacks (FGSM, PGD, and transfer attacks) on the UNSW-NB15 dataset. Through the identification of vulnerable layers, the proposed framework reduces computational complexity and enables the early detection of adversarial samples, thus enhancing the effectiveness and interpretability of adversarial defense mechanisms in NIDS.

In modern GANs, maintaining an Exponential Moving Average (EMA) of the generator's weights is a standard practice, as such an averaged model consistently outperforms the actively trained generator. However, the EMA generator is used for final deployment only and does not influence the training process. To address this missed opportunity, we introduce Self-Distilled GAN (SD-GAN) that employs the EMA generator as a teacher to guide the active generator (student) via perceptual loss. We prove the local asymptotic stability of SD-GAN in the Dirac-GAN setting and show that it dampens the parasitic cycling behavior that plagues the conventional GANs. Empirical evaluations across established architectures and datasets demonstrate that SD-GAN improves the final image quality on several metrics (FID and random-FID in particular), stabilizes the optimization trajectory and provides additional learning guidance that is not trivially correlated with the conventional adversarial loss. It also proves effective for fine-tuning pretrained GAN models.

Current adversarial robustness methods for large language models require extensive datasets of harmful prompts (thousands to hundreds of thousands of examples), yet remain vulnerable to novel attack vectors and distributional shifts. We propose Latent Personality Alignment (LPA), a sample-efficient defense that achieves robustness by training models on abstract personality traits rather than specific harmful behaviors. Using fewer than 100 trait statements and latent adversarial training, LPA achieves comparable attack success rates to methods trained on 150k+ examples, while maintaining superior utility. Critically, LPA generalizes better to unseen attack distributions, reducing misclassification rates by 2.6x compared to baseline across six harm benchmarks -- without ever seeing harmful examples during training. Our results demonstrate that personality-based alignment offers a principled approach to building robust defenses with minimal cost.

Adversarial purification with diffusion models seeks to project adversarial examples back toward the data manifold, but balancing semantic preservation and robustness against adaptive attacks remains challenging. Recent work shows that standard diffusion purification can fail under adaptive evaluation, while test-time score-based optimization is more resilient. Existing optimization defenses, however, typically rely on a single diffusion noise regime or treat timesteps uniformly, overlooking the distinct roles of coarse and fine denoising scales. We propose Temporal Adversarial Rectification Optimization (TARO), an inference-time purification method that builds a temporally guided score prior from multiple denoising views along the diffusion trajectory. TARO forms a coarse-to-fine residual target: high-noise experts provide globally smoothed structure with reduced adversarial sensitivity, while low-noise experts restore image-specific, class-relevant details. A guidance strength controls this temporal correction, allowing TARO to balance robust global rectification with semantic preservation. Empirically, TARO improves robust accuracy across datasets and adaptive threat models in a zero-shot setting, while remaining compatible with complementary adversarial-likelihood objectives for further robustness gains.

Motivational-interviewing-based health coaching is an effective approach for improving mental health and promoting healthy behavior change. However, the scarcity of trained human coaches and the high cost of coaching services make such support inaccessible to many people who could benefit from it. This motivates the development of AI health coaches that can provide scalable and affordable support. Existing methods typically optimize only one side of the interaction: they either train a dialogue agent against a fixed client environment or train a client simulator against a fixed assistant. This one-sided setup can limit exploration of the interaction space and may be inefficient at developing the capabilities required by the target agent and pushing its performance boundaries. In this paper, we propose a dual-agent framework that interactively co-trains both the health coach agent and the client simulator. The coach is optimized with DPO using Pareto-dominant response pairs identified by a multi-dimensional LLM judge. In turn, the client is trained adversarially by reversing these preferences, inducing an implicit adversarial training dynamic. We further show that this co-training process admits a natural stochastic-game interpretation. Extensive experiments demonstrate that our method effectively improves coaching quality across several important dimensions.

Adversarial robustness has been studied extensively for offline deep networks, but less is known about strict single-pass streaming neural learners. This paper studies adversarial robustness in Fuzzy ARTMAP, an Adaptive Resonance Theory architecture based on category competition, complement coding, match tracking, and replay-free prototype updates. We introduce WB-Softmax, a differentiable white-box attack surrogate aligned with ARTMAP's category-competition and map-field prediction mechanism, and formalize a streaming evaluation principle requiring robustness to be assessed on the final deployed model. Across four image benchmarks, WB-Softmax achieves 89-100% attack success on vanilla Fuzzy ARTMAP models. We show that defense rankings can reverse across protocols: offline adversarial training may appear strong under transfer attacks yet collapse under adaptive white-box evaluation, whereas progressive two-stage selective training provides the strongest overall replay-free robustness. We further show that ART's explicit category geometry enables interpretable diagnosis of separation collapse and match-score inversion. These results provide a mechanism-aligned, protocol-aware framework for adversarial robustness in streaming prototype-based learners.

Multimodal recommender systems exploit visual and textual signals to alleviate data sparsity, but this also makes them more vulnerable to evasion-based promotion attacks. Existing defenses are largely limited to single-modal settings and mainly focus on poisoning-based threats, leaving evasion-based threats underexplored. In this work, we first identify a cross-modal gradient mismatch under the multi-user promotion setting, where visual and textual perturbations are optimized in inconsistent directions due to the dominance of distinct user groups. This phenomenon dilutes the attack effectiveness and leads robust training to underestimate worst-case risks. To address this issue, we propose Untargeted Adversarial Training with Multimodal Coordination (UAT-MC). UAT-MC tackles the challenge of unknown targeted items in evasion-based attacks (as opposed to poisoning-based attacks) by treating all items as potential targets, and introduces a gradient alignment mechanism to explicitly correct this mismatch. This design ensures synchronized perturbations across modalities, thereby maximizing adversarial strength for robust training. Extensive experiments demonstrate that UAT-MC significantly improves robustness against promotion attacks while maintaining acceptable recommendation performance under the defense-accuracy trade-off. Code is available at https://github.com/gmXian/UAT-MC.

Large language models (LLMs) remain vulnerable to adversarial prompting despite advances in alignment and safety, often exhibiting harmful behaviors under novel attack strategies. While adversarial training can improve robustness, existing approaches are computationally expensive and difficult to scale. Recent continuous adversarial training methods, such as Continuous adversarial training (CAT) and Continuous Adversarial Preference Optimization (CAPO), address this challenge by leveraging gradient-based perturbations in the embedding space, enabling more efficient and expressive attacks. Building on this paradigm, we propose WARDEN, a distributionally robust adversarial training framework for LLMs that dynamically reweights adversarial examples through an f -divergence ambiguity set around the empirical training distribution. Our method optimizes the worst-case adversarial loss within a divergence ball around the empirical data distribution, automatically emphasizing harder adversarial examples. Using the convex dual formulation, the objective reduces to a log-sum-exp form under the KL divergence, with a dynamical parameter controlling the strength of reweighting. This study leads to a new class of information-theoretic objectives that significantly reduce attack success rates while maintaining model utility. Across multiple LLMs and attack settings, WARDEN substantially reduces attack success rates with computational and utility costs comparable to CAT-, CAPO-, and MixAT-based baselines, making it a practical approach for scalable robust alignment.

Vision-language models (VLMs) are increasingly deployed as trusted authorities -- fact-checking images on social media, comparing products, and moderating content. Users implicitly trust that these systems perceive the same visual content as they do. We show that adversarial examples break this assumption, enabling \emph{AI authority laundering}: an attacker subtly perturbs an image so that the VLM produces confident and authoritative responses about the \emph{wrong} input. Unlike jailbreaks or prompt injections, our attacks do not compromise model alignment; the attack operates entirely at the perceptual level. We demonstrate that standard attacks against publicly available CLIP models transfer reliably to production VLMs -- including GPT-5.4, Claude Opus4.6, Gemini3, and Grok~4.2. Across four attack surfaces, we show that authority laundering can amplify misinformation, disparage individuals, evade content moderation, and manipulate product recommendations. Our attacks have high success rates: In hundreds of attacks targeting identity manipulation and NSFW evasion, we measure success rates of $22 - 100\%$ across six models. No novel attack algorithm is required: basic techniques known for over a decade suffice, establishing a lower bound on attacker capability that should concern defenders. Our results demonstrate that visual adversarial robustness is now a practical -- and still largely unsolved -- safety problem.

Adversarial training is effective on balanced datasets, but its robustness degrades under longtailed class distributions, where tail classes suffer high robust error and unstable decision boundaries. We propose Manifold-Constrained Adversarial Training (MCAT), a unified framework that enforces the semantic validity of adversarial examples by penalizing deviations from class-conditional manifolds in feature space, while promoting balanced geometric separation across classes via an ETF-inspired regularization. We provide theoretical results that link geometric separation to lower bounds on adversarially robust margins, and show that manifold-constrained adversarial risk upperbounds robust risk on high-density semantic regions. Extensive experiments on standard longtailed benchmarks demonstrate consistent improvements in overall, balanced, and tail-class adversarial robustness.

Self-supervised learning methods learn high-quality visual representations, yet recent studies show that these representations often capture demographic biases present in the training data. Existing fairness-aware methods address this by redesigning the self-supervised objective itself, limiting portability across the rapidly evolving landscape of self-supervised learning (SSL) frameworks. We propose ProtoFair, a fairness-aware contrastive loss designed to work alongside existing SSL objectives without modifying them. ProtoFair leverages unsupervised prototype clustering to identify pseudo-counterfactual pairs: samples sharing the same cluster assignment but belonging to different sensitive groups. By pulling these content-matched, cross-group samples together in the embedding space, ProtoFair encourages the encoder to learn representations that are invariant to the sensitive attribute. The method requires only sensitive attribute annotations, no target labels, and integrates seamlessly with both SimCLR and SupCon. Experiments on CelebA and UTKFace demonstrate consistent fairness improvements while maintaining competitive accuracy.

Driving agents trained in one simulated town often perform poorly in a new town because the road shapes, intersections, and lane layouts can be different. This paper studies how to improve this kind of transfer in the CARLA driving simulator without giving the agent any training data from the test towns. The agent is trained only in Town05 and Town06, then evaluated directly in Town03 and Town04. To focus on road-layout differences, all experiments use the same weather and traffic settings. We propose a training method that encourages the agent to learn features that are useful across towns rather than features tied to one training town. During training, the agent is asked to predict the high-level visual meaning of future camera views and is also discouraged from relying on cues that reveal which source town the data came from. These extra learning signals are used only during training; at test time, the driving policy uses the same observation and control interface as the baseline agent. In controlled comparisons with matched DreamerV3-style world-model driving agents, the proposed method achieves the highest mean held-out success: 36.6% on Town03 with a 95% confidence interval of [30.5, 42.7] and 85.6% on Town04 with a 95% confidence interval of [84.0, 87.2], computed across five training seeds. Seed-paired tests against the strongest primary baselines show positive success-rate differences in both held-out towns. Additional experiments show that predicting future visual meaning alone or removing town-specific cues alone is not enough to match the combined method. These results suggest that combining future-scene understanding with reduced reliance on source-town-specific features can improve cross-town driving performance in this CARLA setting.

Vision-language models (VLMs) are increasingly used in autonomous driving because they combine visual perception with language-based reasoning, supporting more interpretable decision-making, yet their robustness to physical adversarial attacks, especially whether such attacks transfer across different VLM architectures, is not well understood and poses a practical risk when attackers do not know which model a vehicle uses. We address this gap with a systematic cross-architecture study of adversarial transferability in VLM-based driving, evaluating three representative architectures (Dolphins, OmniDrive, and LeapVAD) using physically realizable patches placed on roadside infrastructure in both crosswalk and highway scenarios. Our transfer-matrix evaluation shows high cross-architecture effectiveness, with transfer rates of 73-91% (mean TR = 0.815 for crosswalk and 0.833 for highway) and sustained frame-level manipulation over 64.7-79.4% of the critical decision window even when patches are not optimized for the target model.

When instructed to underperform on multiple-choice evaluations, do language models engage with question content or fall back on positional shortcuts? We map the boundary between these regimes using a six-condition adversarial instruction-specificity gradient administered to two instruction-tuned LLMs (Llama-3-8B and Llama-3.1-8B) on 2,000 MMLU-Pro items. Distributional screening (response-position entropy) and an independent content-engagement criterion (difficulty-accuracy correlation) jointly characterise each condition. The gradient reveals three regimes rather than a monotonic transition. Vague adversarial instructions produce moderate accuracy reduction with preserved content engagement. Standard sandbagging and capability-imitation instructions produce positional entropy collapse with partial content engagement. A two-step answer-aware avoidance instruction produces extreme positional collapse, with near-total concentration on a single response position (99.9% and 87.4%) and no measurable content sensitivity. This was the only multi-step instruction tested, and it produced the most extreme shortcut. The attractor position matches each model's content-absent null-prompt default. The effect replicates across both models and four academic domains. Distributional collapse and content engagement can co-occur (50% concordance between screening criteria), indicating that entropy-based screening and difficulty-based content assessment capture partially independent dimensions of response validity. Results suggest that instruction complexity can determine whether adversarial compliance uses content-aware or content-blind mechanisms in small instruction-tuned LLMs under greedy decoding.

Adversarial Training (AT) is one of the most effective methods for developing robust deep neural networks (DNNs). However, AT faces a trade-off problem between clean accuracy and adversarial robustness. In this work, we reveal a surprising phenomenon for the first time: Varying input perturbation intensities for training samples near decision boundaries in AT have minimal impact on model robustness. This finding directly exposes the inconsistency between accuracy and robustness score fluctuations, leading us to identify the misalignment between input and latent spaces as a critical driver of the robustness-accuracy trade-off. To mitigate this misalignment for harmonizing accuracy and robustness, we define Robust Alignment as a new AT target, encouraging the model perception to change with input perturbations provided the final label prediction remains unchanged, which can be achieved via two novel ideas. First, we suggest a reduced and fixed perturbation intensity for those boundary samples, which facilitates the model to utilize the perturbations as learnable patterns, instead of noises that complicate decision boundaries meaninglessly. Second, we propose a Domain Interpolation Consistency Adversarial Regularization (DICAR), based on rigorous theoretical derivations, which explicitly introduces semantic alignment between input and latent spaces into AT. Based on these two ideas, we end up with a new Robust Alignment Adversarial Training (RAAT) method, effectively harmonizing accuracy and robustness. Extensive experiments on CIFAR-10, CIFAR-100, and Tiny-ImageNet with ResNet-18, PreActResNet-18, and WideResNet-28-10 demonstrate the effectiveness of RAAT in improving the trade-off beyond four common baselines and a total of 14 related state-of-the-art (SOTA) works.

The vulnerabilities of deep neural networks against singularities have raised serious concerns regarding their deployment in the physical world. One of the most prominent and impactful physical-world adversarial perturbations is the attachment of patches to clean images, known as an adversarial patch attack. Similarly, natural noises such as Gaussian and Salt&Pepper are highly prevalent in the real world. The current research need arises from the above vulnerabilities and the lack of efforts to tackle these two singularities independently and, especially, in combination. In this research, we have, for the first time, combined these two prominent singularities and proposed a novel dataset. Using this dataset, we have conducted a benchmark study of singularity data-point detection using features from several convolutional neural networks. For classification, rather than the popular neural network-based parameter tuning, we have used traditional yet effective machine learning classifiers. The extensive experiments across various in- and out-of-distribution (OOD) singularities reveal several interesting findings about the effectiveness of classifiers and show that it is hard to defend against adversaries when they are treated independently, and inefficient classifiers are selected.

Knowledge distillation (KD) is a well-known technique to effectively compress a large network (teacher) to a smaller network (student) with little sacrifice in performance. However, most KD methods require a large training set and internal access to the teacher, which are rarely available due to various restrictions. These challenges have originated a more practical setting known as black-box few-shot KD, where the student is trained with few images and a black-box teacher. Recent approaches typically generate additional synthetic images but lack an active strategy to promote their diversity, a crucial factor for student learning. To address these problems, we propose a novel training scheme for generative adversarial networks, where we adaptively select high-confidence images under the teacher's supervision and introduce them to the adversarial learning on-the-fly. Our approach helps expand and improve the diversity of the distillation set, significantly boosting student accuracy. Through extensive experiments, we achieve state-of-the-art results among other few-shot KD methods on seven image datasets. The code is available at https://github.com/votrinhan88/divbfkd.

Fast Adversarial Training (FAT) has attracted significant attention due to its efficiency in enhancing neural network robustness against adversarial attacks. However, FAT is prone to catastrophic overfitting (CO), wherein models overfit to the specific attack used during training and fail to generalize to others. While existing methods introduce diverse hypotheses and propose various strategies to mitigate CO, a systematic and intuitive explanation of CO remains absent. In this work, we innovatively interpret CO through the lens of backdoor. Through validations on pathway division, diverse feature predictions, and universal class distinguishable triggers in CO, we conceptualize CO as a weak trigger variant of unlearnable tasks, unifying CO, backdoor attacks, and unlearnable tasks under a common theoretical framework. Guided by this, we leverage several backdoor inspired strategies to mitigate CO: (i) Recalibrate CO affected model parameters using vanilla fine tuning, linear probing, or reinitialization-based techniques; (ii) Introduce a weight outlier suppression constraint to regulate abnormal deviations in model weights. Extensive experiments support our interpretation of CO and show the efficacy of the proposed mitigation strategies.

Fast Adversarial Training (FAT) has proven effective in enhancing model robustness by encouraging networks to learn perturbation-invariant representations. However, FAT often suffers from catastrophic overfitting (CO), where the model overfits to the training attack and fails to generalize to unseen ones. Moreover, robustness oriented optimization typically leads to notable performance degradation on clean inputs, and such degradation becomes increasingly severe as the perturbation budget grows. In this work, we conduct a comprehensive analysis of how guidance strength affects model performance by modulating perturbation and supervision levels across distinct confidence groups. The findings reveal that low confidence samples are the primary contributors to CO and the robustness accuracy trade off. Building on this insight, we propose a Distribution-aware Dynamic Guidance (DDG) strategy that dynamically adjusts both the perturbation budget and supervision signal. Specifically, DDG scales the perturbation magnitude according to the sample confidence at the ground truth class, thereby guiding samples toward consistent decision boundaries while mitigating the influence of learning spurious correlations. Simultaneously, it dynamically adjusts the supervision signal based on the prediction state of each sample, preventing overemphasis on incorrect signals. To alleviate potential gradient instability arising from dynamic guidance, we further design a weighted regularization constraint. Extensive experiments on standard benchmarks demonstrate that DDG effectively alleviates both CO and the robustness accuracy trade off.

In this work, we study Source-Free Unsupervised Domain Adaptation under corruption-induced domain shifts, where performance degradation is caused by natural image corruptions that go beyond additive noise, including blur, weather effects, and digital artifacts. We propose a diffusion-based, input-level adaptation framework that operates entirely at test time and keeps all source-trained models frozen, explicitly targeting robustness to corrupted target inputs. Our method leverages a source-trained diffusion model as a generative prior and introduces a discriminator-guided adaptive diffusion strategy that dynamically controls the amount of perturbation applied to each test sample. Rather than relying on a fixed diffusion depth, the discriminator determines, on a per-image basis, when sufficient forward diffusion has been applied to suppress corruption-specific artifacts, with each corruption type effectively defining a distinct target domain. This adaptive stopping mechanism applies only the necessary amount of noise to remove domainspecific corruption while preserving class-discriminative structure. The reverse diffusion process then reconstructs a source-aligned image, optionally stabilized through structural guidance, which is classified using a frozen source-trained classifier. We evaluate the proposed approach across a broad spectrum of corruption-induced target domains, covering 15 diverse corruption types, and demonstrate more balanced robustness with competitive or improved performance across non-noise corruptions. Additional analyses reveal how the adaptive diffusion schedule responds to different corruption characteristics, highlighting the practicality, generality, and robustness of the proposed framework. The code is publicly available at https://github.com/fmolivato/dgadiffusion/.

We study whether deep networks for medical imaging learn useful nonrobust features - predictive input patterns that are not human interpretable and highly susceptible to small adversarial perturbations - and how these features impact test performance. We show that models trained only on nonrobust features achieve well above chance accuracy across five MedMNIST classification tasks, confirming their predictive value in-distribution. Conversely, adversarially trained models that primarily rely on robust features sacrifice in-distribution accuracy but yield markedly better performance under controlled distribution shifts (MedMNIST-C). Overall, nonrobust features boost standard accuracy yet degrade out-of-distribution performance, revealing a practical robustness-accuracy trade-off in medical imaging classification tasks that should be tailored to the requirements of the deployment setting.

Facial attribute editing aims to modify target attributes while preserving attribute-irrelevant content and overall image fidelity. Existing GAN-based methods provide favorable controllability, but often suffer from weak alignment between style codes and attribute semantics. Diffusion-based methods can synthesize highly realistic images; however, their editing precision is limited by the entanglement of semantic directions among different attributes. In this paper, we propose AttDiff-GAN, a hybrid framework that combines GAN-based attribute manipulation with diffusion-based image generation. A key challenge in such integration lies in the inconsistency between one-step adversarial learning and multi-step diffusion denoising, which makes effective optimization difficult. To address this issue, we decouple attribute editing from image synthesis by introducing a feature-level adversarial learning scheme to learn explicit attribute manipulation, and then using the manipulated features to guide the diffusion process for image generation, while also removing the reliance on semantic direction-based editing. Moreover, we enhance style-attribute alignment by introducing PriorMapper, which incorporates facial priors into style generation, and RefineExtractor, which captures global semantic relationships through a Transformer for more precise style extraction. Experimental results on CelebA-HQ show that the proposed method achieves more accurate facial attribute editing and better preservation of non-target attributes than state-of-the-art methods in both qualitative and quantitative evaluations.

The growth of agentic AI has drawn significant attention to function calling Large Language Models (LLMs), which are designed to extend the capabilities of AI-powered system by invoking external functions. Injection and jailbreaking attacks have been extensively explored to showcase the vulnerabilities of LLMs to user prompt manipulation. The expanded capabilities of agentic models introduce further vulnerabilities via their function calling interface. Recent work in LLM security showed that function calling can be abused, leading to data tampering and theft, causing disruptive behavior such as endless loops, or causing LLMs to produce harmful content in the style of jailbreaking attacks. This paper introduces a novel function hijacking attack (FHA) that manipulates the tool selection process of agentic models to force the invocation of an attacker-chosen function. While existing attacks focus on semantic preference of the model for function-calling tasks, we show that FHA is largely agnostic to the context semantics and remains effective across domains and function sets. We demonstrate that FHA generalizes to unseen queries and payload perturbations under a fixed target model, reaching 62.5% to 81.9% ASR on held-out queries across 4 function-calling LLMs (instructed and reasoning models), evaluated on the Berkeley Function Calling Leaderboard (BFCL). We further evaluate the cross-model transferability of FHA, showing that FHA can be transferred to other model sizes and families (11.2-27.6% ASR). Our findings further demonstrate the need for strong guardrails and modules for agentic systems.

Fast Adversarial Training (FastAT) seeks to achieve adversarial robustness at a fraction of the computational cost incurred by standard multi-step methods such as PGD-AT. Although numerous FastAT techniques have been proposed in recent years, fair comparison among them remains elusive. Existing benchmarks and public leaderboards typically permit diverse model architectures, varying training configurations, and external data sources, making it unclear whether reported improvements reflect genuine algorithmic advances or merely more favorable experimental conditions. To address this problem, we introduce the FastAT Benchmark, a controlled evaluation framework built on three core design principles: unified architecture requirements, standardized training settings, and strict prohibition of external or synthetic data. The benchmark implements over twenty representative FastAT methods within a single codebase, enabling direct and reproducible comparison. Each method is assessed through a dual-metric evaluation framework that measures both adversarial robustness (accuracy under PGD, AutoAttack, and CR Attack) and computational cost (GPU training time and peak memory footprint). Comprehensive experiments on CIFAR-10, CIFAR-100, and Tiny-ImageNet provide reliable baseline measurements and reveal that well-designed single-step methods can match or surpass PGD-AT robustness at substantially lower cost, while no single method dominates across all evaluation dimensions. The complete benchmark, including source code, configuration files, and experimental results, is publicly available to support transparent and fair evaluation of future FastAT research.

We present RoIt-XMASA, a multilingual dataset that extends the Cross-lingual Multi-domain Amazon Sentiment Analysis to Italian and Romanian, comprising 36,000 labeled reviews across three domains (books, movies, and music) and 202,141 unlabeled samples. To address cross-lingual and cross-domain challenges, we propose a multi-target adversarial training framework that employs loss reversal with meta-learned coefficients to dynamically balance sentiment discrimination with domain and language invariance. XLM-R achieves an F1-score of 66.23% with our approach, outperforming the baseline by 4.64%. Few-shot evaluation shows that Llama-3.1-8B achieves 58.43% F1-score, revealing a meaningful trade-off between the efficiency of prompting-based approaches and the higher performance of task-specific fine-tuning.

While deep learning systems are becoming increasingly prevalent in medical image analysis, their vulnerabilities to adversarial perturbations raise serious concerns for clinical deployment. These vulnerability evaluations largely rely on Attack Success Rate (ASR), a binary metric that indicates solely whether an attack is successful. However, the ASR metric does not account for other factors, such as perturbation strength, perceptual image quality, and cross-architecture attack transferability, and therefore, the interpretation is incomplete. This gap requires consideration, as complex, large-scale deep learning systems, including Vision Transformers (ViTs), are increasingly challenging the dominance of Convolutional Neural Networks (CNNs). These architectures learn differently, and it is unclear whether a single metric, e.g., ASR, can effectively capture adversarial behavior. To address this, we perform a systematic empirical study on four medical image datasets: PathMNIST, DermaMNIST, RetinaMNIST, and CheXpert. We evaluate seven models (VGG-16, ResNet-50, DenseNet-121, Inception-v3, DeiT, Swin Transformer, and ViT-B/16) against seven attack methods at five perturbation budgets, measuring ASR, Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index Measure (SSIM), and $L_2$ perturbation magnitude. Our findings show a consistent pattern: perceptual and distortion metrics are strongly associated with one another and exhibit minimal correlation with ASR. This applies to both CNNs and ViTs. The results demonstrate that ASR alone is an inadequate indicator of adversarial robustness and transferability. Consequently, we argue that a thorough assessment of adversarial risk in medical AI necessitates multi-metric frameworks that encompass not only the attack efficacy but also its methodology and associated overheads.

Adversarial attacks pose a severe threat to the reliability of deep learning models in remote sensing (RS) image classification. Most existing methods rely on direct pixel-wise perturbations, failing to exploit the inherent atmospheric characteristics of RS imagery or survive real-world image degradations. In this paper, we propose FogFool, a physically plausible adversarial framework that generates fog-based perturbations by iteratively optimizing atmospheric patterns based on Perlin noise. By modeling fog formations with natural, irregular structures, FogFool generates adversarial examples that are not only visually consistent with authentic RS scenes but also deceptive. By leveraging the spatial coherence and mid-to-low-frequency nature of atmospheric phenomena, FogFool embeds adversarial information into structural features shared across diverse architectures. Extensive experiments on two benchmark RS datasets demonstrate that FogFool achieves superior performance: not only does it exceed in white-box settings, but also exhibits exceptional black-box transferability (reaching 83.74% TASR) and robustness against common preprocessing-based defenses such as JPEG compression and filtering. Detailed analyses, including confusion matrices and Class Activation Map (CAM) visualizations, reveal that our atmospheric-driven perturbations induce a universal shift in model attention. These results indicate that FogFool represents a practical, stealthy, and highly persistent threat to RS classification systems, providing a robust benchmark for evaluating model reliability in complex environments.

We extend the Weak Adversarial Neural Pushforward Method to the Wigner transport equation governing the phase-space dynamics of quantum systems. The central contribution is a structural observation: integrating the nonlocal pseudo-differential potential operator against plane-wave test functions produces a Dirac delta that exactly inverts the Fourier transform defining the Wigner potential kernel, reducing the operator to a pointwise finite difference of the potential at two shifted arguments. This holds in arbitrary dimension, requires no truncation of the Moyal series, and treats the potential as a black-box function oracle with no derivative information. To handle the negativity of the Wigner quasi-probability distribution, we introduce a signed pushforward architecture that decomposes the solution into two non-negative phase-space distributions mixed with a learnable weight. The resulting method inherits the mesh-free, Jacobian-free, and scalable properties of the original framework while extending it to the quantum setting.

Membership inference attacks (MIAs) threaten the privacy of machine learning models by revealing whether a data point was used during training. Existing MIAs often assume access to public datasets, shadow models, confidence scores or the training distribution, which makes them vulnerable to defenses like confidence masking. Label-only MIAs avoid these assumptions but require thousands of queries per sample. We propose a cost-effective label-only MIA framework based on transferability and model extraction. Querying the target $M$ with active sampling, perturbation-based selection and synthetic data, we extract a surrogate $S$ on which membership inference is performed offline. This shifts query overhead to a one-time extraction phase. It also removes the restriction that defines the label-only setting: the attacker controls $S$ and can read its posteriors and training trajectory, so attacks that cannot be run against $M$ can be run against $S$. On Location, Purchase and Texas, the strongest attack on $S$ improves AUC over the direct attack on $M$ by $0.9$, $5.6$ and $5.0$ percentage points, and improves the true positive rate at $1\%$ false positive rate by $2.8\times$ to $6.3\times$. We characterize how leakage transfer depends on surrogate fidelity, evaluate standard defenses, and report preliminary results on image datasets.

Self-supervised learning (SSL) and diffusion models have respectively advanced representation learning and generative modeling for high-dimensional 3D visual data, yet they are often developed as separate paradigms. Their unification remains challenging under multi-source heterogeneity, as anatomical content must be preserved for analysis while acquisition-related style varies across centers and affects synthesis. In this paper, we propose MeDUET, a 3D Medical image Disentangled UnifiEd PreTraining framework in the variational autoencoder latent space. MeDUET formulates unified pretraining as an empirical factor identifiability problem, aiming to learn domain-invariant content factors for anatomy and domain-specific style factors for appearance. To improve factor separation, MeDUET first uses token demixing with a standard adversarial domain regularizer to establish basic content-style specialization, and further introduces Mixed Factor Token Distillation and Swap-invariance Quadruplet Contrast to reduce mixed-region factor leakage and organize factor spaces with factor-wise invariance and discriminability. With these learned factors, MeDUET transfers effectively to both synthesis and analysis, yielding higher fidelity, faster convergence, and better controllability for synthesis, while achieving competitive or superior domain generalization and label efficiency on diverse datasets, tasks, and modalities. Overall, MeDUET shows that multi-source heterogeneity can serve as useful supervision, with disentanglement providing an effective interface for unifying 3D medical image synthesis and analysis. Our code is available at https://github.com/JK-Liu7/MeDUET.

Diffusion model advances have enabled powerful text-guided image editing, but also raise ethical and legal risks such as deepfakes and unauthorized use. To prevent these risks, adversarial attack-based image immunization has emerged as a promising defense against AI-driven semantic manipulation. Yet, most existing approaches require image-specific optimization or additional neural networks at inference time, hindering scalability and practicality. In this paper, we propose the first universal adversarial perturbation-based image immunization framework that generates a single, image-agnostic adversarial perturbation specifically designed for diffusion-based editing pipelines. Inspired by UAP used in targeted attacks, our method aims to generate a UAP that induces diffusion models to misinterpret the input image as a specific semantic target. Simultaneously, it suppresses original content to misdirect the model's attention during editing, thereby effectively blocking unauthorized edits by overwriting the image's original semantics via the UAP. Extensive experiments show that our method, as the first universal immunization approach, significantly outperforms several baselines in the UAP setting. Notably, despite the inherent difficulty of universal perturbations, our method achieves competitive or superior performance compared to image-specific methods under a more restricted perturbation budget, while also exhibiting strong black-box transferability across diverse diffusion models.

Input transformation-based attacks improve adversarial transferability by aggregating gradients over transformed inputs. Existing analyses mainly explain their efficacy from image diversity, semantic preservation, attention variance or hypothesis space augmentation, yet overlook the critical role of model frontend responses. In this paper, we revisit transformation-based attacks from an implicit ensemble perspective: each transformation can be viewed as a pre-processing operator before the surrogate model, inducing a distinct frontend response for gradient aggregation. Based on this view, we propose FRO, a Frontend Response-Oriented input transformation method that enriches such responses through two complementary operators. The Local Scaling Operator perturbs local content sampling via block-wise stretch-and-shrink operations, while the Projection Operator modifies global spatial organization through coherent perspective deformation. Together, they produce structured transformed views to optimize transferable adversarial perturbations. Experiments on an ImageNet subset show that FRO consistently improves black-box transferability across diverse CNN and Vision Transformer models. We further analyze the effect of implicit ensemble size and evaluate different transformation-based methods under a unified ensemble scale, demonstrating the superiority of designing input transformations from the perspective of front-end response ensembles.

Concept erasure methods aim to remove specific unsafe target concepts in diffusion models while preserving image generation utility. To address the vulnerability that erased concepts can be easily recovered under adversarial attacks, adversarial concept erasure methods integrate adversarial optimization into the concept erasure process. However, existing adversarial concept erasure methods face a trade-off between robustness and computational cost. We attribute this to adversarial optimization techniques that use random samples to approximate the adversarial objective function. Adversarial optimization that uses a small number of samples fails to produce adversarial embeddings that accurately capture the target concept space. To mitigate this limitation, we propose Semantic-Guided Adversarial Optimization, which uses a single sample to produce adversarial embeddings that better capture the target concept space. We also propose Semantic-Guided Concept Erasure, which automatically maps the target concept to a semantically similar surrogate. Extensive experiments on not-safe-for-work content, artistic styles, and object-related concepts demonstrate that our method, S-GRACE (Semantic-Guided Robust Adversarial Concept Erasure) achieves state-of-the-art erasure robustness and superior image generation utility, with significantly lower computational cost than existing methods. Our code is available at https://github.com/Qhong-522/S-GRACE.

Neural operators are widely used as fast surrogates for numerical PDE solvers, mapping input functions to solution functions. However, their generalizability and robustness are not yet clearly defined in the operator-learning setting, which differs from traditional adversarial robustness definitions. This paper studies the generalizability and robustness of a learned neural operator from a solver-integrated perspective, addressing the challenge that the output of a learned operator and a numerical solver tends to change in tandem under input perturbation. First, we formalize the definition of generalization and robustness through a model-solver error operator, identifying fixed-input model-solver loss as generalization metric, and norm-bounded adversarial attack loss increase and Jacobian-error function norm as robustness metric. Second, we identify the solver-integrated adversarial attack as appropriate for PDE operator learning and show why model-only or fixed-ground-truth attacks can be insufficient when the solver output also changes with the input. Third, we develop solver-integrated adversarial training methods for neural operators. Experiments on representative PDE benchmarks show that this solver-integrated adversarial training clearly improves both generalizability and robustness. Deeper solver integration yields more effective attacks, more informative samples, and more efficient training than less integrated alternatives. These results provide a general framework for robust operator training and automatic sample selection without heavy manual intervention. More broadly, the formulation applies to adversarial regression whenever a ground-truth oracle can evaluate, and ideally differentiate, the true input-output map; PDE operator learning is one such case.

Transfer-based adversarial attacks rely on surrogate models to craft perturbations, yet often overfit the surrogate's decision boundary. To address this problem, we propose Inverse Knowledge Distillation (IKD), a simple and attack-agnostic mechanism that maximizes the prediction-distribution discrepancy between benign and adversarial samples on the surrogate model. IKD uses a CE/KL-equivalent soft-label objective to push adversarial predictions away from a fixed benign prediction anchor and enrich the attack with Fisher-sensitive surrogate directions. We prove that, under a matched fixed-anchor implementation, soft-label cross-entropy and KL divergence differ only by a constant entropy term and therefore induce identical gradients, Hessians, and adversarial optimization trajectories. Our information-geometric analysis further derives a quantitative lower bound on dominant Fisher-subspace overlap between surrogate and target models from local same-task stability and a Fisher eigengap, and establishes a sufficient target-margin crossing condition under oriented gradient coherence and target smoothness. This analysis connects IKD's surrogate Fisher sensitivity to cross-model transfer. In contrast, mean squared error uses a different Euclidean pullback in output probability space. IKD integrates seamlessly with standard gradient-based attacks without modifying their optimization pipelines. Extensive ImageNet experiments demonstrate consistent black-box gains across CNN, ViT, and defended models, while ablations confirm CE and KL equivalence and the pronounced disadvantage of MSE. These results establish IKD as an effective and lightweight component for improving adversarial transferability. Code is available at https://github.com/ImmortalTing/IKD.

Large Vision-Language Models (LVLMs) have achieved remarkable performance on multimodal tasks but remain highly vulnerable to small adversarial perturbations in input images. Existing attacks typically target the vision encoder's final output embeddings, implicitly treating the encoder as a uniform attack surface, while a systematic analysis of which internal components are most vulnerable has remained largely unexplored. We show such analysis is essential, as adversarial vulnerability in LVLM vision encoders is structurally concentrated rather than uniformly distributed. Building on this, we propose Vision Encoder Vulnerable-Component-Targeted Universal Adversarial Perturbation (VEV-UAP), a task-agnostic and cost-efficient attack framework. Through a component- and layer-wise analysis of attention mechanisms, we identify the value components in middle layers as critical vulnerabilities that strongly influence downstream language model behavior. VEV-UAP selectively targets these components to generate a single universal perturbation shared across images, without involving textual inputs or the language model during optimization. Experiments across multiple LVLMs and tasks show VEV-UAP achieves state-of-the-art attack success rates with reduced computational overhead. Moreover, a single VEV-UAP transfers across LVLMs sharing the same vision encoder, even when paired with different language models, making it a practical framework for scalable robustness evaluation.

Generative flow networks (GFlowNets) are sequential sampling models trained to match a given distribution. GFlowNets have been successfully applied to various structured object generation tasks, sampling a diverse set of high-reward objects quickly. We propose expected flow networks (EFlowNets), which extend GFlowNets to stochastic environments. We show that EFlowNets outperform other GFlowNet formulations in stochastic tasks such as protein design. We then extend the concept of EFlowNets to adversarial environments, proposing adversarial flow networks (AFlowNets) for two-player zero-sum games. We show that AFlowNets learn to find above 80% of optimal moves in Connect-4 via self-play and outperform AlphaZero in tournaments.

Transfer learning is a critical technique that enables the application of knowledge gained from existing tasks or domains to improve performance on a new one, reducing the need for extensive data and training in each new context. Many existing transfer learning methods rely on leveraging information from source populations closely resembling the target population. However, this approach often overlooks valuable knowledge that may be present in different yet potentially related auxiliary samples. When dealing with a limited amount of target data and multiple source data, we introduce a novel approach, Guided Adversarial Robust Transfer (GART) learning, that breaks free from strict similarity constraints. GART is designed to optimize the most adversarial loss with respect to a collection of source mixture distributions that guarantee excellent prediction performances for the target data. We establish the closed form of the population GART and show that the GART estimator achieves a faster convergence rate than the model fitted with the target data. Our simulation studies suggest that GART outperforms existing transfer learning methods, attaining higher robustness and accuracy. We highlight GART's predictiveness and robustness by applying it to form genetic prediction models of high-density lipoprotein cholesterol using multi-institutional biobank-linked electronic health records data.

Quantitative assessment of muscle MRI is crucial for monitoring neuromuscular disorders (NMD). This study introduces an automated radiomic phenotyping framework based on original features engineered across five main architectural domains: quantitative morphometry, spatial distribution, geometric shape, interactions between progressive fat replacement stages, and graph-based topology. Utilizing 1184 MRI scans from the CoMPaSS-NMD project, we map the complex 3D architecture of heterogeneous intramuscular lipodegeneration into objective, morphologically interpretable biomarkers. We introduce a graph-based skeletonization of fat infiltrates to quantify muscle architectural changes, establishing a multi-dimensional extension of traditional, spatially-agnostic volume metrics by mapping topological networks across the entire 3D muscle volume. Statistical screening via non-parametric Kruskal-Wallis analysis confirmed the discriminative power of these novel descriptors across the genetic hierarchy. Notably, topological network metrics (e.g., SF1_Skel_Nodes, $\epsilon^2$ = 0.2656) and interface dynamics metrics (e.g., SF2_To_SF1_Dist_Min, $\epsilon^2$ = 0.2092) demonstrated substantial effect sizes, providing deeper structural insights than classical volumetric assessments. Post-hoc pairwise evaluations and UMAP projections further indicated the capability of these topological and 3D geometric invariants to capture disease-specific macroscopic infiltration patterns. These results demonstrate that global architectural features represent a highly promising class of biomarkers for differential diagnosis, offering new avenues for tracking longitudinal disease dynamics in neuromuscular diagnostics. The developed automated feature extraction pipeline is integrated and available within the MUSCAT (MUSCle fAt Topology) library.