超越线性表示假设：文本到图像模型中的非线性激活引导
原标题：Beyond the Linear Representation Hypothesis: Non-Linear Activation Steering in Text-to-Image Models
AI 摘要
该研究挑战了机械可解释性中常用的线性表示假设（LRH），指出文本到图像扩散模型中的视觉概念（如天气、昆虫变态）在激活空间中并非沿直线遍历，而是呈现非线性曲线轨迹。作者提出 KANSteer，利用 Kolmogorov-Arnold 网络（KAN）将概念遍历建模为一条穿过中间状态的一维曲线，使引导方向随概念变化而调整，同时保持可解释性。实验表明，在多个概念和扩散 Transformer 上，激活轨迹显著偏离直线，KANSteer 比线性引导能更准确地拟合中间属性并实现更平滑的遍历与更忠实的外推。
正文节选
Beyond the Linear Representation Hypothesis: Non-Linear Activation Steering in Text-to-Image Models Abstract Mechanistic interpretability often relies on the Linear Representation Hypothesis (LRH), which assumes that high-level concepts are encoded as linear directions in activation space. Yet a natural visual concept does not necessarily require a linear visual transition: between sunny and stormy lies an intermediate weather state such as a sky with a few white clouds, not simply a weaker stor