Detection Is Cheap, Routing Is Learned: Why Refusal-Based Alignment Evaluation Fails
Paper • 2603.18280 • Published • 1
Greg Frank PRO
gregfrank
·
AI & ML interests
Alignment, mechanistic interpretability, model behavior, agentic red-teaming
Recent Activity
authored a paper about 4 hours ago
Detection Is Cheap, Routing Is Learned: Why Refusal-Based Alignment Evaluation Fails authored a paper about 5 hours ago
How Alignment Routes: Localizing, Scaling, and Controlling Policy Circuits in Language Models updated a collection 1 day ago
alignmentOrganizations
None yet
authored a
paper about 4 hours ago
authored a
paper about 5 hours ago
updated a
collection 1 day ago
submitted a
paper to Daily Papers 1 day ago