Huazhen Lin

Unsupervised Transfer Learning via Adversarial Contrastive Training

Adv-SSL: Adversarial Self-Supervised Representation Learning with Theoretical Guarantees

AMSC: Adaptive Multi-Dimensional Structured Compression with Theoretical Guarantees

Olica: Efficient Structured Pruning of Large Language Models without Retraining

Self-supervised Transfer Learning via Adversarial Contrastive Training

Estimation error of gradient descent in deep regressions