{
  "id": 6559,
  "url": "https://arxiv.org/abs/2603.29410v1",
  "title": "AGFT: Alignment-Guided Fine-Tuning for Zero-Shot Adversarial Robustness of Vision-Language Models",
  "summary": "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. Un",
  "authors": "Yubo Cui, Xianchao Guan, Zijun Xiong, Zheng Zhang",
  "category": "research",
  "topics": "safety-alignment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-03-31T08:17:27.000Z",
  "fetched_at": "2026-07-14T16:32:33.103Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/6559",
  "original_url": "https://arxiv.org/abs/2603.29410v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}