1) EMNLP 2022 Train Flat, Then Compress
2) ESAM PPT
3) B站：SAM算法有何用 清华 温凯越
4) Improving Sharpness-Aware Minimization with Fisher Mask for Better Generalization on Language Models
5) Adversarial Weight Perturbation Helps Robust Generalization
6) NeurIPS 2022 sharpness_aware_training_for_free
7) ICML 2021 (ASAM)Adaptive SAM for Scale-Invariant Learning of Deep Neural Networks
8) EMNLP 2022 Improving Sharpness-Aware Minimization with Fisher Mask
9) CVPR 2022 LookSAM
