5 papers
Global-Local Attention Decomposition for Terrain Encoding in Humanoid Perceptive Locomotion
Shengcheng Fu, Yang Zhang, Zhanxiang Cao +4
Although reinforcement learning has significantly advanced humanoid locomotion, perceptive policies still struggle on sparse-foothold terrain and constrained environments. Success…
UniLab: A Heterogeneous Architecture for Robot RL Beyond GPU-Dominant Paradigms
Yufei Jia, Zhanxiang Cao, Mingrui Yu +48
Simulation-based RL for contemporary robot control is increasingly organized around GPU-resident simulation: physics, rollout collection, and learning are placed on a single GPU-ce…
HierKick: Hierarchical Reinforcement Learning for Vision-Guided Soccer Robot Control
Yizhi Chen, Zheng Zhang, Zhanxiang Cao +7
Controlling soccer robots involves multi-time-scale decision-making, which requires balancing long-term tactical planning and short-term motion execution. Traditional end-to-end re…
'1'-bit Count-based Sorting Unit to Reduce Link Power in DNN Accelerators
Ruichi Han, Yizhi Chen, Tong Lei +2
Interconnect power consumption remains a bottleneck in Deep Neural Network (DNN) accelerators. While ordering data based on '1'-bit counts can mitigate this via reduced switching a…
Late Breaking Results: Quamba-SE: Soft-edge Quantizer for Activations in State Space Models
Yizhi Chen, Ahmed Hemani
We propose Quamba-SE, a soft-edge quantizer for State Space Model (SSM) activation quantization. Unlike existing methods, using standard INT8 operation, Quamba-SE employs three ada…