Revisiting Over-refusal in Safety-aligned Large Language Models: A Representation and Optimization Perspective
arXiv:2610.39802v1 Announce Type: new Abstract: Over-refusal is a critical side effect of safety alignment in large language models (LLMs), causing harmless but superficially risky queries to be mistakenly rejected. Despite the effectiveness of existing mitigation methods, why safety-aligned LLMs may still reject harmless queries under correct safety supervision remains insufficiently understood. To address this gap, we analyze over-refusal through the lens of representation and optimization. W…