Max Robotics — Global Service Robot Directory + US Certification

About

Siao Liu is a rising researcher in robotics and artificial intelligence, with a focus on advancing robotic manipulation through reinforcement learning and lifelong learning systems. Their work centers on developing algorithms that enable robots to acquire, retain, and generalize skills over extended periods—a critical challenge for real-world autonomy. Liu’s most notable contribution is the paper "DiffSkill: Improving Reinforcement Learning through Diffusion-Based Skill Denoiser for Robotic Manipulation" (2024), which has garnered 11 citations. This work introduces a novel diffusion-based denoising mechanism that enhances reinforcement learning by refining skill policies, leading to more robust and adaptable robotic manipulation. Building on this, Liu’s 2025 paper "Think Small, Act Big: Primitive Prompt Learning for Lifelong Robot Manipulation" tackles the persistent problem of catastrophic forgetting in lifelong learning. By proposing a primitive prompt learning framework, Liu enables robots to leverage prior knowledge for continuous skill acquisition without performance degradation. With a growing citation impact and a focus on scalable, memory-efficient solutions, Siao Liu is contributing to the next generation of autonomous robots capable of learning and adapting throughout their operational lifetimes.