Ming Hu

One Bad Sample May Spoil the Whole Batch: A Novel Backdoor-Like Attack Towards Large Batch Processing

Enhancing Zeroth-Order Fine-Tuning for LLMs via Gradient-Guided Subspace Selection

Rising from Ashes: Generalized Federated Learning via Dynamic Parameter Reset

One Arrow, Two Hawks: Sharpness-aware Minimization for Federated Learning via Global Model Trajectory

Enhancing Zeroth-Order Fine-Tuning for LLMs via Gradient-Guided Subspace Selection

PerFIT: Personalized Federated Instruction Tuning via Neural Architecture Search