generalization behavior
- Algorithm- and Data-Dependent Generalization Bounds for Diffusion Models
- Better Language Model Inversion by Compactly Representing Next-Token Distributions
- Beyond Benign Overfitting in Nadaraya-Watson Interpolators
- DLoFT: Gradient-Decoupled Fine-Tuning for Generalizable Long Chain-of-Thought Reasoning
- Ineq-Comp: Benchmarking Human-Intuitive Compositional Reasoning in Automated Theorem Proving of Inequalities
- On the VC dimension of deep group convolutional neural networks