通过探索输入流形揭示Transformer的感知机制
Unveiling Transformer Perception by Exploring Input Manifolds
AI总结：
该研究提出一种基于输入流形序列形变理论的Transformer输入空间等价类探索方法，可实现同类与异类实例检索，并支持结果的人类可读化解释。
摘要：
本文提出了一种用于探索Transformer模型输入空间中等价类的通用方法。该方法基于严谨的数学理论，将Transformer架构的内部层描述为输入流形的序列形变。通过对模型雅可比矩阵在输出空间上定义的距离度量的拉回进行特征分解，我们能够重构输入空间中的等价类并在其间导航。我们的方法支持两种互补的探索流程：第一种检索与原始实例产生相同类别概率分布的输入实例，从而识别同一等价类内的元素；第二种则发现产生不同类别概率分布的实例，有效实现向不同等价类的导航。最后，我们展示了如何通过将检索到的实例的嵌入投影回人类可读格式，对其进行有意义的解释。
英文摘要：
This paper introduces a general method for the exploration of equivalence classes in the input space of Transformer models. The proposed approach is based on sound mathematical theory which describes the internal layers of a Transformer architecture as sequential deformations of the input manifold. Using eigendecomposition of the pullback of the distance metric defined on the output space through the Jacobian of the model, we are able to reconstruct equivalence classes in the input space and navigate across them. Our method enables two complementary exploration procedures: the first retrieves input instances that produce the same class probability distribution as the original instance-thus identifying elements within the same equivalence class-while the second discovers instances that yield a different class probability distribution, effectively navigating toward distinct equivalence classes. Finally, we demonstrate how the retrieved instances can be meaningfully interpreted by projecting their embeddings back into a human-readable format.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
