2 papers
cs.LG2026
Manifold of Failure: Behavioral Attraction Basins in Language Models
Sarthak Munshi, Manish Bhatt, Vineeth Sai Narajala +4
While prior work has focused on projecting adversarial examples back onto the manifold of natural data to restore safety, we argue that a comprehensive understanding of AI safety r…
cs.CR2026
The Defense Trilemma: Why Prompt Injection Defense Wrappers Fail?
Manish Bhatt, Sarthak Munshi, Vineeth Sai Narajala +6
We prove that no continuous, utility-preserving wrapper defense-a function that preprocesses inputs before the model sees them-can make all outputs strictly safe for a…