Han Shu

6

Papers

39

Total Citations

Papers (6)

TinySAM: Pushing the Envelope for Efficient Segment Anything Model

ViSpec: Accelerating Vision-Language Models with Vision-Aware Speculative Decoding

NeurIPS 2025arXiv

ExCP: Extreme LLM Checkpoint Compression via Weight-Momentum Joint Shrinking

Frequency Domain Compact 3D Convolutional Neural Networks

Optical Flow Distillation: Towards Efficient and Stable Video Style Transfer

Adder Attention for Vision Transformer