通过高频滤波提升模型泛化能力，让深度网络更关注通用特征。
Deep Edge Filter: Return of the Human-Crafted Layer in Deep Learning
- 用高通滤波分离特征中的高频通用信息与低频领域偏见
- 在视觉、文本、3D、音频任务中均实现稳定性能提升
- 无需改动网络结构，适配多种模型和数据模态
我们提出 Deep Edge Filter，一种对深度神经网络特征进行高通滤波的新方法，以提升模型泛化能力。该方法基于假设：神经网络在深层特征的高频分量中编码任务相关语义信息，而低频分量则存储领域特定偏见。通过将原始特征减去低通滤波输出，该方法有效分离出可泛化的表示，同时保持网络结构完整性。在视觉、文本、3D 和音频等多个领域的实验表明，无论模型架构或数据模态如何，该方法均能持续提升性能。分析显示，该方法诱导特征稀疏化，并有效分离高频成分，为核心假设提供了实证支持。代码已公开于 https://github.com/dongkwani/DeepEdgeFilter。
原文摘要 · Abstract (English)
We introduce the Deep Edge Filter, a novel approach that applies high-pass filtering to deep neural network features to improve model generalizability. Our method is motivated by our hypothesis that neural networks encode task-relevant semantic information in high-frequency components while storing domain-specific biases in low-frequency components of deep features. By subtracting low-pass filtered outputs from original features, our approach isolates generalizable representations while preserving architectural integrity. Experimental results across diverse domains such as Vision, Text, 3D, and Audio demonstrate consistent performance improvements regardless of model architecture and data modality. Analysis reveals that our method induces feature sparsification and effectively isolates high-frequency components, providing empirical validation of our core hypothesis. The code is available at https://github.com/dongkwani/DeepEdgeFilter.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。