Yuan Zhang

YOLOv5-Multibackbone-Compression. YOLOv5 Series Multi-backbone(TPH-YOLOv5, Ghostnet, ShuffleNetv2, Mobilenetv3Small, EfficientNetLite, PP-LCNet, SwinTransformer YOLO), Module(CBAM, DCN), Pruning (EagleEye, Network Slimming), Quantization (MQBench) and Deployment (TensorRT, ncnn) Compression Tool Box.

SparseVLMs. [ICML'25] Official implementation of paper "SparseVLM: Visual Token Sparsification for Efficient Vision-Language Model Inference".

DVS-gestureRecognition. Official implementation of paper "Razor SNN: Efficient Spiking Neural Network with Temporal Embeddings", ICANN 2023.

FreeKD. [CVPR'24] Official implementation of paper "FreeKD: Knowledge Distillation via Semantic Frequency Prompt".

AvatarKD. [ACM MM'23] Official implementation of paper "Avatar Knowledge Distillation: Self-ensemble Teacher Paradigm with Uncertainty".

MasKD. Official implementation of paper "Masked Distillation with Receptive Tokens", ICLR 2023.

SEED. Official implementation of paper "SEED: Targeted Data Selection by Weighted Independent Set".

DAMO-YOLO. DAMO-YOLO: a fast and accurate object detection method with some new techs, including NAS backbones, efficient RepGFPN, ZeroHead, AlignedOTA, and distillation enhancement.

AutoV. [ECCV'26] Official implementation of paper "AutoV: Loss-Oriented Ranking for Visual Prompt Retrieval in LVLMs".