Mingjia Shi

4 papers University of Virginia

• Drag-and-Drop LLMs: Zero-Shot Prompt-to-Weights
• Pruning-Robust Mamba with Asymmetric Multi-Scale Scanning Paths
• REPA Works Until It Doesn’t: Early-Stopped, Holistic Alignment Supercharges Diffusion Training
• Tackling Feature-Classifier Mismatch in Federated Learning via Prompt-Driven Feature Transformation