Navigating the Trade-off: A Synthesis of Defensive Strategies for Zero-Shot Adversarial Robustness in Vision-Language Models
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918116776214528 |
|---|---|
| author | Xu, Zane Sun, Jason |
| author_facet | Xu, Zane Sun, Jason |
| contents | This report synthesizes eight seminal papers on the zero-shot adversarial robustness of vision-language models (VLMs) like CLIP. A central challenge in this domain is the inherent trade-off between enhancing adversarial robustness and preserving the model's zero-shot generalization capabilities. We analyze two primary defense paradigms: Adversarial Fine-Tuning (AFT), which modifies model parameters, and Training-Free/Test-Time Defenses, which preserve them. We trace the evolution from alignment-preserving methods (TeCoA) to embedding space re-engineering (LAAT, TIMA), and from input heuristics (AOM, TTC) to latent-space purification (CLIPure). Finally, we identify key challenges and future directions including hybrid defense strategies and adversarial pre-training. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_05237 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Navigating the Trade-off: A Synthesis of Defensive Strategies for Zero-Shot Adversarial Robustness in Vision-Language Models Xu, Zane Sun, Jason Computer Vision and Pattern Recognition Artificial Intelligence This report synthesizes eight seminal papers on the zero-shot adversarial robustness of vision-language models (VLMs) like CLIP. A central challenge in this domain is the inherent trade-off between enhancing adversarial robustness and preserving the model's zero-shot generalization capabilities. We analyze two primary defense paradigms: Adversarial Fine-Tuning (AFT), which modifies model parameters, and Training-Free/Test-Time Defenses, which preserve them. We trace the evolution from alignment-preserving methods (TeCoA) to embedding space re-engineering (LAAT, TIMA), and from input heuristics (AOM, TTC) to latent-space purification (CLIPure). Finally, we identify key challenges and future directions including hybrid defense strategies and adversarial pre-training. |
| title | Navigating the Trade-off: A Synthesis of Defensive Strategies for Zero-Shot Adversarial Robustness in Vision-Language Models |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2508.05237 |