Weiming Hu

5 papers Institute of automation, Chinese academy of science

• Each Complexity Deserves a Pruning Policy
• MI-TRQR: Mutual Information-Based Temporal Redundancy Quantification and Reduction for Energy-Efficient Spiking Neural Networks
• Online Segment Any 3D Thing as Instance Tracking
• SynCL: A Synergistic Training Strategy with Instance-Aware Contrastive Learning for End-to-End Multi-Camera 3D Tracking
• Yggdrasil: Bridging Dynamic Speculation and Static Runtime for Latency-Optimal Tree-Based LLM Decoding