Jinxuan Xu

Rutgers Sexual and Reproductive Health and Rights, Nanjing University of Aeronautics and Astronautics

Papers

Total Citations

H-Index

About

Jinxuan Xu is a rising researcher at the forefront of integrating large language models (LLMs) with robotic manipulation. His work focuses on bridging the gap between high-level reasoning and low-level physical actions, enabling robots to understand and execute tasks with greater autonomy. His most impactful contribution is the development of **RT-Grasp**, a novel framework that uses reasoning tuning within a multimodal LLM to enable robots to grasp objects based on implicit human intent, not just explicit commands. This work, published in 2024, has already garnered 12 citations, signaling its significant influence on the field. Xu further refined this concept in his subsequent paper, "Reasoning Grasping via Multimodal Large Language Model" (4 citations), which deepens the exploration of context-aware robotic action. Beyond LLM-based reasoning, Xu has also contributed to spatial perception with his work on **Real-Time Plane Detection with Consistency from Point Cloud Sequences** (5 citations), addressing the critical challenge of fast and reliable environment mapping for real-time robotic tasks. By combining advanced reasoning with robust perception, Jinxuan Xu is pioneering a new generation of intelligent, context-aware robots.