MLP-like models built entirely upon multi-layer perceptrons have recently been revisited, exhibiting the comparable performance with transformers. It is one of most promising architectures due to the excellent trade-off between network capability and efficiency in the large-scale recognition tasks. However, its generalization performance to heterogeneous tasks is inferior to other architectures (e.g., CNNs and transformers) due to the extensive retention of domain information. To address this problem, we propose a novel frequency-aware MLP architecture, in which the domain-specific features are filtered out in the transformed frequency domain, augmenting the invariant descriptor for label prediction. Specifically, we design an adaptive Fourier filter layer, in which a learnable frequency filter is utilized to adjust the amplitude distribution by optimizing both the real and imaginary parts. A low-rank enhancement module is further proposed to rectify the filtered features by adding the low-frequency components from SVD decomposition. Finally, a momentum update strategy is utilized to stabilize the optimization to fluctuation of model parameters and inputs by the output distillation with weighted historical states. To our best knowledge, we are the first to propose a MLP-like backbone for domain generalization. Extensive experiments on three benchmarks demonstrate significant generalization performance, outperforming the state-of-the-art methods by a margin of 3%, 4% and 9%, respectively.
Spatial-Frequency-Aware Implicit Neural Representation of Multidimensional Signals via MLP-KAN Fusion
Institute of Computational Imaging, Beijing Information Science and Technology University（北京信息科技大学计算成像研究所）；College of Computer Science (College of Software), Inner Mongolia University（内蒙古大学计算机学院（软件学院））；Research Center for Spatiotemporal Intelligence, Inner Mongolia University（内蒙古大学时空智能研究中心）
CtrlNeRF: The Generative Neural Radiation Fields for the Controllable Synthesis of High-fidelity 3D-Aware Images
School of Computer and Artificial Intelligence, Zhengzhou University（郑州大学计算机与人工智能学院）；Department of Electrical and Computer Engineering, California State Polytechnic University（加州州立理工大学波莫纳分校电气与计算机工程学院）
Improving Cross-domain Few-shot Classification with Multilayer Perceptron
Institute of Artificial Intelligence and Robotics, Xi'an Jiaotong University（西安交通大学人工智能与机器人研究所）；Westlake University Institute of Advanced Technology, Westlake Institute for Advanced Study（西湖大学先进技术研究院）
SETA: Semantic-Aware Token Augmentation for Domain Generalization
National Institute of Healthcare Data Science, Nanjing University（南京大学国家健康医疗大数据研究院）；School of Computer Science and Engineering, Southeast University（东南大学计算机科学与工程学院）
Sharpness-Aware Gradient Matching for Domain Generalization
The Hong Kong Polytechnic University（香港理工大学）；Center for Artificial Intelligence and Robotics（人工智能与机器人中心）；HKISI CAS（中国科学院香港创新研究院）；State Key Laboratory of Multimodal Artificial Intelligence Systems（多模态人工智能系统全国重点实验室）；CASIA（中国科学院自动化研究所）
