Skip to content
NewA calmer, faster ZeroHour

Region-Level Black-Box Defense Against Stealthy Embedding-Space Backdoors in CLIP

arXiv cs.CRby Ahmed Abdelnaby ()1 min read2 · 1 read
AI summary · grok-4.7Summaries by Vani AI

CLIPGuard localizes and inpaints CLIP embedding backdoors, cutting attack success to about 1% in black-box tests.

CLIPGuard is a lightweight black-box defense against embedding-space backdoors in CLIP that does not need model weights, gradients, logits, or clean validation data. It measures segment-wise embedding perturbations and semantically inpaints only suspicious regions. On STL-10, ImageNet, and triggers including BadCLIP, BadNets, blended, patch, and typographic attacks, attack success fell to as low as 1.05% while clean accuracy reached 86.34%, beating CleanCLIP and CleanerCLIP.

Key facts
  • Targets stealthy embedding-space backdoors rather than logit triggers.
  • Black-box: no weights, gradients, logits, or clean validation data.
  • Inpaints only image segments with suspicious embedding shifts.
  • Attack success as low as 1.05% with clean accuracy up to 86.34%.
Explain this AI news with Vani AIWhat was announced, what is new and why it matters, in a new chat

From the original article

1 min read at arxiv.org

Contrastive Language--Image Pretraining (CLIP) has emerged as a dominant vision backbone due to its strong transferability and zero-shot capabilities. However, recent studies reveal a critical vulnerability: embedding-space backdoor attacks. By poisoning only a tiny fraction of image--text pairs, adversaries can implant stealthy triggers that induce targeted shifts in CLIP's joint embedding space.

arXiv cs.CR

By Ahmed Abdelnaby · arxiv.org

Read the full article at arXiv cs.CR

A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.