AI.info

Hoang Phan

Explore Hoang Phan on AI.info.

• Sharpening Tax in Post-Training
• Hyperparameter Transfer Enables Consistent Gains of Matrix-Preconditioned Optimizers Across Scales