Focusing on zero-shot classification, this study demonstrates that 3D vision-language models exhibit heightened sensitivity to small coordinate perturbations, highlighting the need for a more rigorous security evaluation of 3D vision-language models.
Abstract
Pre-trained 3D vision-language models have demonstrated strong semantic generalization and robustness to distribution shifts. However, the implications of semantic robustness for geometric stability remain unclear. This study revisits the adversarial robustness of 3D vision-language models when confronted with adversarial point clouds. Focusing on zero-shot classification, this study demonstrates that these models exhibit heightened sensitivity to small coordinate perturbations. The behavior of adversarial perturbations is further analyzed under widely used point-cloud preprocessing mechanisms, revealing that naive filtering or reconstruction mainly suppresses irregular perturbations produced by vanilla gradient-based attacks and provides limited protection against stronger attack methods. To this end, a refined adversarial objective is introduced with two complementary priors that encourage adversarial point clouds to remain smooth and geometrically plausible: a statistical prior that regularizes the sampling distribution, and a geometric prior that promotes consistency with a plausible object-surface manifold. These findings highlight the need for a more rigorous security evaluation of 3D vision-language models.
The recent progress in the vision-language model (VLM) research made it one of the key aspects of artificial intelligence due to joint vision and language processing capabilities required for such tasks as image retrieval, visual question answering, autonomous systems and medical image analysis. However, despite the excellent results obtained with such systems, they still appear to be extremely vulnerable to adversarial perturbations which result in a significant decline in prediction accuracy and reliability due to even subtle input transformations. The purpose of this paper is to provide a systematic method for analyzing the adversarial robustness of vision-language models under both clean and perturbed conditions. The suggested framework is based on Semantic Counterfactual Augmentation (SCA) and Curriculum Contrastive Adversarial Training (CCAT) and aims to maintain semantic consistency and increase the robustness of the model respectively. The framework is tested using a CLIP-based vision-language model on the subset of Flickr8k dataset with Fast Gradient Sign Method (FGSM) and Projected Gradient Descent (PGD) attacks. As a result of the experiment, the performance gap is identified between the clean and perturbed images which confirms the vulnerabilities of the conventional VLM while indicating better robustness consistency with the suggested framework.
Kumari Anjali, Karnatakam Veda Sahithi, Sakinala Jyotsna· 2026 4th International Confe...· 0 citations
While Large Vision-Language Models (LVLMs), represented by LLaVA and GPT-4V, have demonstrated remarkable capabilities, their visual inputs remain vulnerable to adversarial attacks, posing significant security risks. Existing defense methods predominantly target single-task scenarios (e.g., zero-shot classification) and consequently lack generalizability across various multimodal tasks. To address this limitation, we propose a dual adversarial fine-tuning framework that jointly optimizes visual and semantic supervision signals from two modalities, enhancing model robustness while generalizing across multiple downstream tasks. The proposed framework comprises two core components, i.e., $\textbf{Visual}$ supervision branch and $\textbf{Semantic}$ supervision branch. The former branch leverages features from clean images, extracted via a frozen original vision encoder, to guide adversarial robustness while the latter incorporates caption-image alignment as a contextual signal to preserve semantic coherence under attack. Moreover, our method achieves cross-task robustness by simply replacing the CLIP vision encoder in the original model, with no need of separate task-specific retraining or architecture modifications.Extensive experiments demonstrate that our approach outperforms the state-of-the-art method in adversarial robustness evaluation across zero-shot classification, image captioning, and visual question answering (VQA) tasks.
Sibo Wang, Jie Zhang, Shiguang Shan et al.· 0 citations
Vision-language pre-trained models (VLPs) are widely used in real-world applications. However, they remain vulnerable to adversarial attacks. Although adversarial detection methods have demonstrated success in single-modality settings (either vision or language), their effectiveness and reliability in multimodal models such as VLPs remain largely unexplored. In this work, we study the geometry of VLP embedding spaces and observe structured anisotropy that differs from unimodal vision models. Our theoretical analysis shows that under this anisotropic structure, adversarial attacks increase the expected geometric separation between clean and adversarial examples (AEs). Specifically, we demonstrate that AEs consistently exhibit greater expected distances to randomly sampled points than their clean counterparts, indicating that AEs tend to push representations out of manifold regions. Building on these insights, we propose GeoDetect, which leverages these off-manifold deviations via geometric scores to identify AEs. Through comprehensive evaluations, we show that our approach reliably detects AEs across diverse VLP architectures and threat settings, covering unimodal and multimodal attacks as well as adaptive attacks, thereby providing a robust and practical approach to improving the safety and reliability of these models.
Afsaneh Hasanebrahimi, Hanxun Huang, Christopher Leckie et al.· 0 citations
This work proposes a white-box spectral-subspace-guided attack (SSGRA) that aligns intermediate representations with the subspace spanned by the bottom right singular vectors, offering a spectral interpretation of adversarial vulnerability in VLMs, providing insights for improving their robustness.
Chethan Krishnamurthy Ramanaik, Tobias Callies, Michael Hecht et al.· 0 citations
This work proposes RITA, a Robust test-tIme prompt-TAdaptation framework that shifts from sample-level estimates to distribution-level alignment, and employs optimal transport to align the distribution of augmented visual features with textual prototypes, mitigating adversarial outliers and rectifying cross-modal semantic misalignment.
Xingyu Zhu, Huanshen Wu, Shuo Wang et al.· 1 citation
LiDAR semantic segmentation is a key perception task in autonomous driving, where false predictions can affect downstream planning and safety-critical decision-making. Although adversarial attacks, and specifically adversarial examples, have been widely studied for image classification and 3D point cloud segmentation, unrestricted adversarial examples remain largely unexplored in the space of 2D range images, which are projections of 3D point clouds. The proposed method is, to the best of our knowledge, the first diffusion-based unrestricted adversarial attack against 2D range-image segmentation, using adversarial guidance from a segmentation loss. By applying guidance directly during sampling, the method produces unrestricted adversarial examples that remain close to the learned LiDAR data manifold while inducing structured segmentation errors. Experiments on the SemanticKITTI dataset using RangeNet++ and CENet segmentation networks demonstrate that the attack provides adjustable degradation across guidance strengths and transfers across segmentation architectures. Compared with norm-bounded FGSM and SegPGD baselines, the proposed attack offers a distinct effectiveness-realism trade-off, achieving controllable white-box and transfer degradation while maintaining competitive distributional and visual realism.
Stavros Bouras, Antonios Makris, A. Gkillas et al.· 0 citations