1
Faithfulness to Refusal: A Causal Audit of Neuron Selectors
用因果审计拆解神经元选择器，探究模型拒绝行为是否忠实可靠，值得深读。
arXiv:2607.05355v1 Announce Type: cross Abstract: Attribution scores increasingly identify which neuron rows of a language model matter for applicatio…