Spiking Neural Networks (SNNs) are energy-efficient and biologically plausible, ideal for embedded and security-critical systems, yet their adversarial robustness remains open. Existing adversarial attacks often overlook SNNs' bio-plausible dynamics. We propose Spike-PTSD, a biologically inspired adversarial attack framework modeled on abnormal neural firing in Post-Traumatic Stress Disorder (PTSD). It localizes decision-critical layers, selects neurons via hyper/hypoactivation signatures, and optimizes adversarial examples with dual objectives. Across six datasets, three encoding types, and four models, Spike-PTSD achieves over 99% success rates, systematically compromising SNN robustness. Code: https://github.com/bluefier/Spike-PTSD.

Recent advances in GAN and diffusion models have significantly improved the realism and controllability of facial deepfake manipulation, raising serious concerns regarding privacy, security, and identity misuse. Proactive defenses attempt to counter this threat by injecting adversarial perturbations into images before manipulation takes place. However, existing approaches remain limited in effectiveness due to suboptimal perturbation injection strategies and are typically designed under white-box assumptions, targeting only simple GAN-based attribute editing. These constraints hinder their applicability in practical real-world scenarios. In this paper, we propose AEGIS, the first diffusion-guided paradigm in which the AdvErsarial facial images are Generated for Identity Shielding. We observe that the limited defense capability of existing approaches stems from the peak-clipping constraint, where perturbations are forcibly truncated due to a fixed $L_\infty$-bounded. To overcome this limitation, instead of directly modifying pixels, AEGIS injects adversarial perturbations into the latent space along the DDIM denoising trajectory, thereby decoupling the perturbation magnitude from pixel-level constraints and allowing perturbations to adaptively amplify where most effective. The extensible design of AEGIS allows the defense to be expanded from purely white-box use to also support black-box scenarios through a gradient-estimation strategy. Extensive experiments across GAN and diffusion-based deepfake generators show that AEGIS consistently delivers strong defense effectiveness while maintaining high perceptual quality. In white-box settings, it achieves robust manipulation disruption, whereas in black-box settings, it demonstrates strong cross-model transferability.

Vision-language-action (VLA) models have shown strong performance in robotic manipulation, yet their robustness to physically realizable adversarial attacks remains underexplored. Existing studies reveal vulnerabilities through language perturbations and 2D visual attacks, but these attack surfaces are either less representative of real deployment or limited in physical realism. In contrast, adversarial 3D textures pose a more physically plausible and damaging threat, as they are naturally attached to manipulated objects and are easier to deploy in physical environments. Bringing adversarial 3D textures to VLA systems is nevertheless nontrivial. A central obstacle is that standard 3D simulators do not provide a differentiable optimization path from the VLA objective function back to object appearance, making it difficult to optimize through an end-to-end manner. To address this, we introduce Foreground-Background Decoupling (FBD), which enables differentiable texture optimization through dual-renderer alignment while preserving the original simulation environment. To further ensure that the attack remains effective across long-horizon and diverse viewpoints in the physical world, we propose Trajectory-Aware Adversarial Optimization (TAAO), which prioritizes behaviorally critical frames and stabilizes optimization with a vertex-based parameterization. Built on these designs, we present Tex3D, the first framework for end-to-end optimization of 3D adversarial textures directly within the VLA simulation environment. Experiments in both simulation and real-robot settings show that Tex3D significantly degrades VLA performance across multiple manipulation tasks, achieving task failure rates of up to 96.7\%. Our empirical results expose critical vulnerabilities of VLA systems to physically grounded 3D adversarial attacks and highlight the need for robustness-aware training.

Recent Tracking-by-Query-Propagation (TBP) methods have advanced Multi-Object Tracking (MOT) by enabling end-to-end (E2E) pipelines with long-range temporal modeling. However, this reliance on query propagation introduces unexplored architectural vulnerabilities to adversarial attacks. We present FADE, a novel attack framework designed to exploit these specific vulnerabilities. FADE employs two attack strategies targeting core TBP mechanisms: (i) Temporal Query Flooding: Generates spurious temporally consistent track queries to exhaust the tracker's limited query budget, forcing it to terminate valid tracks. (ii) Temporal Memory Corruption: Directly attacks the query updater's memory by severing temporal links via state de-correlation and erasing the learned feature identity of matched tracks. Furthermore, we introduce a differentiable pipeline to optimize these attacks for physical-world realizability by leveraging simulations of advanced perception sensor spoofing. Experiments on MOT17 and MOT20 benchmarks demonstrate that FADE is highly effective against state-of-the-art TBP trackers, causing significant identity switches and track terminations.

DNNs are susceptible to defects like backdoors, adversarial attacks, and unfairness, undermining their reliability. Existing approaches mainly involve retraining, optimization, constraint-solving, or search algorithms. However, most methods rely on gradient calculations, restricting applicability to specific activation functions (e.g., ReLU), or use search algorithms with uninterpretable localization and repair. Furthermore, they often lack generalizability across multiple properties. We propose SHARPEN, integrating interpretable fault localization with a derivative-free optimization strategy. First, SHARPEN introduces a Deep SHAP-based localization strategy quantifying each layer's and neuron's marginal contribution to erroneous outputs. Specifically, a hierarchical coarse-to-fine approach reranks layers by aggregated impact, then locates faulty neurons/filters by analyzing activation divergences between property-violating and benign states. Subsequently, SHARPEN incorporates CMA-ES to repair identified neurons. CMA-ES leverages a covariance matrix to capture variable dependencies, enabling gradient-free search and coordinated adjustments across coupled neurons. By combining interpretable localization with evolutionary optimization, SHARPEN enables derivative-free repair across architectures, being less sensitive to gradient anomalies and hyperparameters. We demonstrate SHARPEN's effectiveness on three repair tasks. Balancing property repair and accuracy preservation, it outperforms baselines in backdoor removal (+10.56%), adversarial mitigation (+5.78%), and unfairness repair (+11.82%). Notably, SHARPEN handles diverse tasks, and its modular design is plug-and-play with different derivative-free optimizers, highlighting its flexibility.

Deep neural networks have demonstrated excellent performance in SAR target detection tasks but remain susceptible to adversarial attacks. Existing SAR-specific attack methods can effectively deceive detectors; however, they often introduce noticeable perturbations and are largely confined to digital domain, neglecting physical implementation constrains for attacking SAR systems. In this paper, a novel Adversarial Attenuation Patch (AAP) method is proposed that employs energy-constrained optimization strategy coupled with an attenuation-based deployment framework to achieve a seamless balance between attack effectiveness and stealthiness. More importantly, AAP exhibits strong potential for physical realization by aligning with signal-level electronic jamming mechanisms. Experimental results show that AAP effectively degrades detection performance while preserving high imperceptibility, and shows favorable transferability across different models. This study provides a physical grounded perspective for adversarial attacks on SAR target detection systems and facilitates the design of more covert and practically deployable attack strategies. The source code is made available at https://github.com/boremycin/SAAP.

Transformer-based NLP models remain vulnerable to adversarial perturbations, yet existing repair methods face a fundamental trade-off: gradient-based approaches offer flexibility but lack verifiability and often overfit; methods that do provide repair guarantees are restricted to the final layer or small networks, significantly limiting the parameter search space available for repair. We present WARP (Weight-Adjusted Repair with Provability), a constraint-based repair framework that extends repair beyond the last layer of Transformer models. WARP formulates repair as a convex quadratic program derived from a first-order linearization of the logit gap, enabling tractable optimization over a high-dimensional parameter space. Under the condition that the first-order approximation holds, this formulation induces three per-sample guarantees: (i) a positive margin constraint ensuring correct classification on repaired inputs, (ii) preservation constraints over a designated remain set, and (iii) a certified robustness radius derived from Lipschitz continuity. To ensure feasibility across varying model architectures, we introduce a sensitivity-based preprocessing step that conditions the optimization landscape accordingly. We further show that the iterative optimization procedure converges to solutions satisfying all repair constraints under mild assumptions. Empirical evaluation on encoder-only Transformers with varying layer architectures validates that these guarantees hold in practice while improving robustness to adversarial inputs. Our results demonstrate that guaranteed, generalizable Transformer repair is achievable through principled constraint-based optimization.

World models -- learned internal simulators of environment dynamics -- are rapidly becoming foundational to autonomous decision-making in robotics, autonomous vehicles, and agentic AI. Yet this predictive power introduces a distinctive set of safety, security, and cognitive risks. Adversaries can corrupt training data, poison latent representations, and exploit compounding rollout errors to cause catastrophic failures in safety-critical deployments. World model-equipped agents are more capable of goal misgeneralisation, deceptive alignment, and reward hacking precisely because they can simulate the consequences of their own actions. Authoritative world model predictions further foster automation bias and miscalibrated human trust that operators lack the tools to audit. This paper surveys the world model landscape; introduces formal definitions of trajectory persistence and representational risk; presents a five-profile attacker capability taxonomy; and develops a unified threat model extending MITRE ATLAS and the OWASP LLM Top 10 to the world model stack. We provide an empirical proof-of-concept on trajectory-persistent adversarial attacks (GRU-RSSM: A_1 = 2.26x amplification, -59.5% reduction under adversarial fine-tuning; stochastic RSSM proxy: A_1 = 0.65x; DreamerV3 checkpoint: non-zero action drift confirmed). We illustrate risks through four deployment scenarios and propose interdisciplinary mitigations spanning adversarial hardening, alignment engineering, NIST AI RMF and EU AI Act governance, and human-factors design. We argue that world models must be treated as safety-critical infrastructure requiring the same rigour as flight-control software or medical devices.

The primary tools used to monitor and defend object detectors under adversarial attack assume that when accuracy degrades, detection count drops in tandem. This coupling was assumed, not measured. We report a counterexample observed on a single model: under standard PGD, EMS-YOLO, a spiking neural network (SNN) object detector, retains more than 70% of its detections while mAP collapses from 0.528 to 0.042. We term this count-preserving accuracy collapse Quality Corruption (QC), to distinguish it from the suppression that dominates untargeted evaluation. Across four SNN architectures and two threat models (l-infinity and l-2), QC appears only in one of the four detectors tested (EMS-YOLO). On this model, all five standard defense components fail to detect or mitigate QC, suggesting the defense ecosystem may rely on a shared assumption calibrated on a single substrate. These results provide, to our knowledge, the first evidence that adversarial failure modes can be substrate-dependent.

Vision-language models (VLMs) are vulnerable to adversarial image perturbations. Existing works based on adversarial training against task-specific adversarial examples are computationally expensive and often fail to generalize to unseen attack types. To address these limitations, we introduce Paraphrase-Decomposition-Aggregation (PDA), a training-free defense framework that leverages text augmentation to enhance VLM robustness under diverse adversarial image attacks. PDA performs prompt paraphrasing, question decomposition, and consistency aggregation entirely at test time, thus requiring no modification on the underlying models. To balance robustness and efficiency, we instantiate PDA as invariants that reduce the inference cost while retaining most of its robustness gains. Experiments on multiple VLM architectures and benchmarks for visual question answering, classification, and captioning show that PDA achieves consistent robustness gains against various adversarial perturbations while maintaining competitive clean accuracy, establishing a generic, strong and practical defense framework for VLMs during inference.

This work presents a robust multi-class classification framework for handwritten digits that combines diffusion-driven feature denoising with a hybrid feature representation. Inspired by our previous work on brain tumor classification, the proposed approach operates in a feature space to improve the robustness to noise and adversarial attacks. First, the input images are converted into tight, interpretable exemplification using Nonnegative Matrix Factorization (NNMF). In parallel, special deep features are extracted using a computational neural network (CNN). These integral features are combined into a united hybrid representation. To improve robustness, a step diffusion operation is used in the feature space by gradually adding Gaussian noise. A feature denoiser network is trained to reverse this operation and rebuild clean representations from tilted inputs. The courteous features are then applied for multi-class classification. The suggested method is evaluated in both baseline and adversarial settings using AutoAttack. The experimental outcome present that the diffusion-based hybrid model is both effective and robust, the CNN baseline models outperforming while maintain powerful classification performance. These results explain the activity of feature-level diffusion defense for reliable multi-class handwritten digit classification.

Multimodal large-language models (MLLMs) often experience degraded safety alignment when harmful queries exploit cross-modal interactions. Models aligned on text alone show a higher rate of successful attacks when extended to two or more modalities. In this work, we propose a simple conditional decoding strategy, CASA (Classification Augmented with Safety Attention) that utilizes internal representations of MLLMs to predict a binary safety token before response generation. We introduce a novel safety attention module designed to enhance the model's ability to detect malicious queries. Our design ensures robust safety alignment without relying on any external classifier or auxiliary head, and without the need for modality-specific safety fine-tuning. On diverse benchmarks such as MM-SafetyBench, JailbreakV-28k, and adversarial audio tests, CASA lowers the average attack success rate by more than 97% across modalities and across attack types. Our empirical evaluations also show that CASA maintains strong utility in benign inputs, a result validated through both automated and human evaluations (via 13 trained annotators). Together, these results highlight CASA as a simple and generalizable framework to improve multimodal LLM safety.

Although multimodal large language models (MLLMs) are increasingly deployed in real-world applications, their instruction-following behavior leaves them vulnerable to prompt injection attacks. Existing prompt injection methods predominantly rely on textual prompts or perceptible visual prompts that are observable by human users. In this work, we study imperceptible visual prompt injection against powerful closed-source MLLMs, where adversarial instructions are embedded in the visual modality. Our method adaptively embeds the malicious prompt into the input image via a bounded text overlay to provide semantic guidance. Meanwhile, the imperceptible visual perturbation is iteratively optimized to align the feature representation of the attacked image with those of the malicious visual and textual targets at both coarse- and fine-grained levels. Specifically, the visual target is instantiated as a text-rendered image and progressively refined during optimization to more faithfully represent the desired semantics and improve transferability. Extensive experiments on two multimodal understanding tasks across multiple closed-source MLLMs demonstrate the superior performance of our approach compared to existing methods.

Autonomous AI agents are being deployed with filesystem access, email control, and multi-step planning. This thesis contributes to four open problems in AI safety: understanding dangerous internal computations, removing dangerous behaviors once embedded, testing for vulnerabilities before deployment, and predicting when models will act against deployers. ACDC automates circuit discovery in transformers, recovering all five component types from prior manual work on GPT-2 Small by selecting 68 edges from 32,000 candidates in hours rather than months. Latent Adversarial Training (LAT) removes dangerous behaviors by optimizing perturbations in the residual stream to elicit failure modes, then training under those perturbations. LAT solved the sleeper agent problem where standard safety training failed, matching existing defenses with 700x fewer GPU hours. Best-of-N jailbreaking achieves 89% attack success on GPT-4o and 78% on Claude 3.5 Sonnet through random input augmentations. Attack success follows power law scaling across text, vision, and audio, enabling quantitative forecasting of adversarial robustness. Agentic misalignment tests whether frontier models autonomously choose harmful actions given ordinary goals. Across 16 models, agents engaged in blackmail (96% for Claude Opus 4), espionage, and actions causing death. Misbehavior rates rose from 6.5% to 55.1% when models stated scenarios were real rather than evaluations. The thesis does not fully resolve any of these problems but makes each tractable and measurable.

Adversarial robustness evaluation faces a critical challenge as new defense paradigms emerge that can exploit limitations in existing assessment methods. This paper reveals that Dummy Classes-based defenses, which introduce an additional "dummy" class as a safety sink for adversarial examples, achieve significantly overestimated robustness under conventional evaluation strategies like AutoAttack. The fundamental limitation stems from these attacks' singular focus on misleading the true class label, which aligns perfectly with the defense mechanism--successful attacks are simply captured by the dummy class. To address this gap, we propose Dummy-Aware Weighted Attack (DAWA), a novel evaluation method that simultaneously targets both the true label and dummy label with adaptive weighting during adversarial example synthesis. Extensive experiments demonstrate that DAWA effectively breaks this defense paradigm, reducing the measured robustness of a leading Dummy Classes-based defense from 58.61% to 29.52% on CIFAR-10 under l_infty perturbation (epsilon=8/255). Our work provides a more reliable benchmark for evaluating this emerging class of defenses and highlights the need for continuous evolution of robustness assessment methodologies.

Pre-trained vision-language models (VLMs) exhibit strong zero-shot generalization but remain vulnerable to adversarial perturbations. Existing classification-guided adversarial fine-tuning methods often disrupt pre-trained cross-modal alignment, weakening visual-textual correspondence and degrading zero-shot performance. In this paper, we propose an Alignment-Guided Fine-Tuning (AGFT) framework that enhances zero-shot adversarial robustness while preserving the cross-modal semantic structure. Unlike label-based methods that rely on hard labels and fail to maintain the relative relationships between image and text, AGFT leverages the probabilistic predictions of the original model for text-guided adversarial training, which aligns adversarial visual features with textual embeddings via soft alignment distributions, improving zero-shot adversarial robustness. To address structural discrepancies introduced by fine-tuning, we introduce a distribution consistency calibration mechanism that adjusts the robust model output to match a temperature-scaled version of the pre-trained model predictions. Extensive experiments across multiple zero-shot benchmarks demonstrate that AGFT outperforms state-of-the-art methods while significantly improving zero-shot adversarial robustness.

The integration of multimodal models into Presentation Attack Detection (PAD) for ID Documents represents a significant advancement in biometric security. Traditional PAD systems rely solely on visual features, which often fail to detect sophisticated spoofing attacks. This study explores the combination of visual and textual modalities by utilizing pre-trained multimodal models, such as Paligemma, Llava, and Qwen, to enhance the detection of presentation attacks on ID Documents. This approach merges deep visual embeddings with contextual metadata (e.g., document type, issuer, and date). However, experimental results indicate that these models struggle to accurately detect PAD on ID Documents.

The global Lipschitz constant of a neural network governs both adversarial robustness and generalization. Conventional approaches to ``certified training" typically follow a train-then-verify paradigm: they train a network and then attempt to bound its Lipschitz constant. Because the efficient ``trivial bound" (the product of the layerwise Lipschitz constants) is exponentially loose for arbitrary networks, these approaches must rely on computationally expensive techniques such as semidefinite programming, mixed-integer programming, or branch-and-bound. We propose a different paradigm: rather than designing complex verifiers for arbitrary networks, we design networks to be verifiable by the fast trivial bound. We show that directly penalizing the trivial bound during training forces it to become tight, thereby effectively regularizing the true Lipschitz constant. To achieve this, we identify three structural obstructions to a tight trivial bound (dead neurons, bias terms, and ill-conditioned weights) and introduce architectural mitigations, including a novel notion of norm-saturating polyactivations and bias-free sinusoidal layers. Our approach avoids the runtime complexity of advanced verification while achieving strong results: we train robust networks on MNIST with Lipschitz bounds that are small (orders of magnitude lower than comparable works) and tight (within 10% of the ground truth). The experimental results validate the theoretical guarantees, support the proposed mechanisms, and extend empirically to diverse activations and non-Euclidean norms.

Deep Neural Networks (DNNs) achieve strong performance in semantic segmentation for robotic perception but remain vulnerable to adversarial attacks, threatening safety-critical applications. While robustness has been studied for image classification, semantic segmentation in robotic contexts requires specialized architectures and detection strategies.

Face morphing attacks compromise biometric security by creating document images that verify against multiple identities, posing significant risks from document issuance to border control. Differential Morphing Attack Detection (D-MAD) offers an effective countermeasure, particularly when employing face demorphing to disentangle identities blended in the morph. However, existing methods lack operational generalizability due to limited training data and the assumption that all document inputs are morphs. This paper presents SFDemorpher, a framework designed for the operational deployment of face demorphing for D-MAD that performs identity disentanglement within joint StyleGAN latent and high-dimensional feature spaces. We introduce a dual-pass training strategy handling both morphed and bona fide documents, leveraging a hybrid corpus with predominantly synthetic identities to enhance robustness against unseen distributions. Extensive evaluation confirms state-of-the-art generalizability across unseen identities, diverse capture conditions, and 13 morphing techniques, spanning both border verification and the challenging document enrollment stage. Our framework achieves superior D-MAD performance by widening the margin between the score distributions of bona fide and morphed samples while providing high-fidelity visual reconstructions facilitating explainability.