Weimin Lyu

Unrolled Networks are Conditional Probability Flows in MRI Reconstruction

Backdooring CLIP through Concept Confusion

Concept-Guided Backdoor Attack on Vision Language Models

BadConcepts: Backdooring VLMs with Visual Concepts

Bézier Meets Diffusion: Robust Generation Across Domains for Medical Image Segmentation

ImpScore: A Learnable Metric For Quantifying The Implicitness Level of Sentences

Editable Concept Bottleneck Models

Geometry of Long-Tailed Representation Learning: Rebalancing Features for Skewed Distributions

Backdooring Vision-Language Models with Out-Of-Distribution Data

Representation Confusion: Towards Representation Backdoor on CLIP via Concept Activation