over-smoothing
A phenomenon in deep learning, particularly in graph neural networks, where increased layers lead to loss of distinct feature representations, making different samples indistinguishable. Addressing over-smoothing is essential for maintaining model performance.
- Conditional Diffusion Anomaly Modeling on Graphs
- D2SA: Dual-Stage Distribution and Slice Adaptation for Efficient Test-Time Adaptation in MRI Reconstruction
- Effects of Dropout on Performance in Long-range Graph Learning Tasks
- Frequency-Aware Token Reduction for Efficient Vision Transformer
- Generative Graph Pattern Machine
- How Particle System Theory Enhances Hypergraph Message Passing
- MultiNet: Adaptive Multi-Viewed Subgraph Convolutional Networks for Graph Classification
- OCN: Effectively Utilizing Higher-Order Common Neighbors for Better Link Prediction
- Wavy Transformer