Dynamic Mixture of Progressive Parameter-Efficient Expert Library for Lifelong Robot Learning
Text2World: Benchmarking Large Language Models for Symbolic World Model Generation
Zeroth-Order Actor-Critic: An Evolutionary Framework for Sequential Decision Problems
FREA: Feasibility-Guided Generation of Safety-Critical Scenarios with Reasonable Adversariality
Performance-Driven Controller Tuning via Derivative-Free Reinforcement Learning
Background
Research interests include embodied AI, reinforcement learning, robotic control, and autonomous driving. The goal is to develop efficient algorithms that enable autonomous systems to solve complex tasks in the real world.