paper-with-me

홈 › Papers

AGC: Adaptive Geodesic Correction for Adversarial Robustness on Vision-Language Models

2026-05-15 · Zhiwei Li, Jiacheng Xue, Weining Wang, Ajian Liu, Xingyu Gao, Zhenan Sun, Qi Li arxiv

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.

📄 PDF Abstract BibTeX arXiv:2605.15584

Code (0)

등록된 구현이 없습니다.

Tasks

Adversarial RobustnessData Augmentation

Similar Papers 제목 키워드 기반

Adversarial attacks on neural networks through canonical Riemannian foliations

2022-03-02 · Eliot Tron, Nicolas Couellan, Stéphane Puechmorel

Deep learning models are known to be vulnerable to adversarial attacks. Adversarial learning is therefore becoming a crucial task. We propose a new vision on neural network robustness using Riemannian geometry and foliat…

Adversarial Attack

Hydra: An Agentic Reasoning Approach for Enhancing Adversarial Robustness and Mitigating Hallucinations in Vision-Language Models

2025-04-19 · Chung-En, Yu, Hsuan-Chih, Chen 외

To develop trustworthy Vision-Language Models (VLMs), it is essential to address adversarial robustness and hallucination mitigation, both of which impact factual accuracy in high-stakes applications such as defense and …

Adversarial AttackAdversarial DefenseAdversarial RobustnessHallucination+1

GGSS: Geodesic-Gated Spherical Steering for Inference-Time Debiasing of Generative Vision-Language Models

2026-08-26 · Yiqun Sun, Junyu Chen, Pengfei Wei, Lawrence B. Hsieh arxiv

Generative vision-language models (VLMs) are increasingly used in human-centered settings, yet they can produce demographically biased outputs even when images differ only in controlled attributes such as perceived race …

Boosting Accuracy and Robustness of Student Models via Adaptive Adversarial Distillation

2023-01-01 · CVPR 2023 1 · Bo Huang, Mingyang Chen, Yi Wang, Junda Lu 외

Distilled student models in teacher-student architectures are widely considered for computational-effective deployment in real-time applications and edge devices. However, there is a higher risk of student models to …

Adversarial RobustnessKnowledge Distillation

Counterfactual Explanations on Robust Perceptual Geodesics

2026-01-26 · Eslam Zaher, Maciej Trzaskowski, Quan Nguyen, Fred Roosta arxiv

Latent-space optimization methods for counterfactual explanations - framed as minimal semantic perturbations that change model predictions - inherit the ambiguity of Wachter et al.'s objective: the choice of distance met…