Sparse Autoencoders as Plug-and-Play Firewalls for Adversarial Attack Detection in VLMs
Paper • 2605.07447 • Published • 3
Duke Sun
dukesun99
AI & ML interests
None yet
Recent Activity
upvoted a paper about 9 hours ago
Sparse Autoencoders as Plug-and-Play Firewalls for Adversarial Attack Detection in VLMs submitted a paper 25 days ago
Don't Retrieve, Navigate: Distilling Enterprise Knowledge into Navigable Agent Skills for QA and RAGOrganizations
None yet
upvoted a paper about 9 hours ago