删掉AI脑子里的一个概念，还不伤别的
想让AI忘掉某个概念（比如性别、种族），但保留其他信息，很难——因为概念之间常纠缠。这篇论文发现：AI的表示其实集中在低维的“流形”上，就像数据挤在一个弯曲的面上。于是他们提出：删除操作应该沿着这个面走，而不是乱砍。方法叫MANCE：先估计出这个流形，然后每次删除更新都投影回流形上。在119个设置（文本+图像）上测试，效果比之前好，且不伤其他信息。它不是你明天能用上的，但给AI安全对齐提供了新思路：尊重数据本身的几何结构，而不是暴力修改。
📄 原文摘要(英文)
Concept erasure aims to remove a target concept from a representation while preserving the other information encoded in it. This is difficult because representations encode many concepts that are often correlated with the erasure target, so removing the target risks damaging them. We propose the Manifold Constraint Hypothesis (MCH): if natural representations concentrate on a structured, lower-dimensional manifold, then interventions should be constrained to that manifold and better preserve other information encoded in the representation during interventions. We instantiate MCH in a new concept erasure method: MANifold aware Concept Erasure (MANCE). MANCE performs iterative updates to the representations using signals from a classifier that predicts a target concept. We estimate the manifold using representations obtained from natural inputs, and then we project the concept removal update to the estimated manifold. We perform extensive evaluation on 119 settings spanning text and vision, including 13 language models, three NLP concepts, and 40 CelebA-CLIP attributes. Employing MANCE on top of previous methods shows consistent improved leakage results. We also introduce MANCE+ and MANCE++, which prepend a closed-form erasure algorithm before employing MANCE, achieving better leakage--surgicality tradeoffs relative to matched full-space updates. MANCE++, our best method, achieves state-of-the-art results on nonlinear concept erasure. These results support MCH in the erasure setting: interventions should be constrained to the natural representation manifold.