Test-Time Attention Purification for Backdoored Large Vision Language Models
Test-Time Attention Purification for Backdoored Large Vision Language Models explores CleanSight offers a novel, training-free defense against backdoor attacks in large vision-language models by purifying inputs at test time.. Commercial viability score: 7/10 in Security in AI.
BUILDER'S SANDBOX
Build This Paper
Use an AI coding agent to implement this research.
OpenAI CodexAI Agent
Lightweight coding agent in your terminal.
Claude CodeAI Agent
Agentic coding tool for terminal workflows.
AntiGravity IDEScaffolding
AI agent mindset installer and workflow scaffolder.
CursorIDE
AI-first code editor built on VS Code.
VS CodeIDE
Free, open-source editor by Microsoft.
Recommended Stack
Startup Essentials
Estimated $10K - $14K over 6-10 weeks.
See exactly what it costs to build this -- with 3 comparable funded startups.
7-day free trial. Cancel anytime.
Discover the researchers behind this paper and find similar experts.
7-day free trial. Cancel anytime.
Related Papers
- Builds On ThisRobustness of Vision Language Models Against Split-Image Harmful Input AttacksViability: 5.0
- Builds On ThisRisk Awareness Injection: Calibrating Vision-Language Models for Safety without Compromising UtilityViability: 6.0
- Builds On ThisThe Trigger in the Haystack: Extracting and Reconstructing LLM Backdoor TriggersViability: 2.0
- Builds On ThisJailbreaks on Vision Language Model via Multimodal ReasoningViability: 3.0
- Prior WorkASAP: Attention-Shift-Aware Pruning for Efficient LVLM Inference