Qiyang Li

Papers

Total Citations

H-Index

About

Qiyang Li is a robotics and machine learning researcher whose work spans reinforcement learning, multi-agent coordination, and generalizable robot policy design. His early research drew inspiration from biological swarm intelligence to develop distributed coordination policies for large-scale robotic swarms, addressing the fundamental challenge of enabling collective, adaptive behavior through local interactions alone — work that has garnered attention within the swarm robotics community. More recently, Li has turned his focus toward the frontier of generalist robot learning, contributing to RLDG (Robotic Generalist Policy Distillation via Reinforcement Learning), a framework that bridges specialist reinforcement learning agents and broad generalist policies such as OpenVLA and Octo. By leveraging RL-trained specialists to generate high-quality fine-tuning datasets, RLDG meaningfully improves generalization to unseen scenarios and enables composition across long-horizon tasks — a significant step toward robust, deployable robot intelligence. Though early in citation accumulation, Li's research trajectory reflects a consistent commitment to scalable, adaptive robotic systems, positioning him as an emerging voice in the intersection of deep reinforcement learning and real-world robotics deployment.