Wenhan Yang

Quantization with Purpose: Loss-Aware Bit Allocation for Gradient Compression

Time Is All It Takes: Spike-Retiming Attacks on Event-Driven Spiking Neural Networks

MTL-UE: Learning to Learn Nothing for Multi-Task Learning

End-to-End Low-Light Enhancement for Object Detection with Learned Metadata from RAWs

Prompt-Guided Alignment with Information Bottleneck Makes Image Compression Also a Restorer

Privacy-Shielded Image Compression: Defending Against Exploitation from Vision-Language Pretrained Models

Which Tasks Should Be Compressed Together? A Causal Discovery Approach for Efficient Multi-Task Representation Compression