NeurIPS 2025 Explorer
Concepts
Authors
Glossary
johnsanterre.github.io
test performance
4 papers
Ascent Fails to Forget
Benign Overfitting in Single-Head Attention
How Does Label Noise Gradient Descent Improve Generalization in the Low SNR Regime?
Optimal Control for Transformer Architectures: Enhancing Generalization, Robustness and Efficiency