Amos: An Adam-style Optimizer with Adaptive Weight Decay towards Model-Oriented Scale Amos:一种面向模型的自适应权重衰减的Adam风格优化器.
Contrastive Learning Rivals Masked Image Modeling in Fine-tuning via Feature Distillation 特征蒸馏使对比学习在微调时击败了掩码图像建模.