Junjie Wen

Papers

Total Citations

H-Index

About

Junjie Wen is an emerging researcher at the forefront of robotic manipulation and embodied artificial intelligence, with a particular focus on bridging natural language understanding with physical robot control. His work tackles some of the most fundamental challenges in robotics: enabling machines to interpret human instructions and translate them into precise, context-aware actions across diverse and complex tasks. Wen's most notable contribution, "Language-Conditioned Robotic Manipulation with Fast and Slow Thinking" (2024, 10 citations), draws inspiration from dual-process cognitive theory to develop systems capable of handling tasks ranging from simple pick-and-place operations to sophisticated intent recognition and visual reasoning. This work reflects a creative synthesis of cognitive science principles with cutting-edge robotics, demonstrating his interdisciplinary approach to research. His more recent work on disentangled action spaces addresses the critical challenge of multimodal action distributions in multi-task settings, introducing discrete policy frameworks that improve generalization across varied manipulation scenarios. Though early in his career, Wen's research agenda positions him as a promising voice in the growing field of language-guided embodied AI, with contributions that speak directly to the goal of making robots more intuitive, flexible, and capable collaborators.