Tracing the Dynamics of Refusal: Exploiting Latent Refusal Trajectories for Robust Jailbreak Detection
Paper • 2605.02958 • Published • 1
huxulion
huxulion
AI & ML interests
None yet
Recent Activity
upvoted a paper about 1 month ago
Tracing the Dynamics of Refusal: Exploiting Latent Refusal Trajectories for Robust Jailbreak DetectionOrganizations
None yet
upvoted a paper about 1 month ago