Li Shen

Papers

Total Citations

H-Index

About

Li Shen is an emerging researcher at the intersection of reinforcement learning, deep learning architectures, and continual learning. His work focuses on advancing the capabilities of artificial intelligence systems in sequential decision-making and adaptive learning environments, with a particular emphasis on integrating modern neural network architectures into reinforcement learning frameworks. Shen's most notable contribution is his comprehensive survey on transformer-based reinforcement learning, "On Transforming Reinforcement Learning With Transformers: The Development Trajectory" (2024), which has rapidly accumulated 60 citations, reflecting the strong community interest in this timely synthesis. By systematically mapping how transformers — architectures that revolutionized natural language processing and computer vision — can be leveraged for reinforcement learning, Shen has provided researchers with an invaluable roadmap for the field's trajectory. His more recent work, "Continual Diffuser (CoD): Mastering Continual Offline RL With Experience Rehearsal" (2025), tackles the challenging problem of continual learning in offline reinforcement learning settings using diffusion-based models, addressing a critical real-world limitation of static training paradigms. Together, these contributions position Shen as a promising voice in advancing robust, adaptable AI systems for complex, dynamic environments.