Max Robotics — Global Service Robot Directory + US Certification

About

Jiatao Zhang is a rising researcher at the forefront of robotic task planning, where their work bridges large language models (LLMs) and autonomous decision-making. Their primary research areas include long-horizon task planning, multi-level decomposition, and the application of open-source LLMs in robotics. Zhang’s major contributions center on developing methods that enable robots to execute complex, sequential tasks with greater reliability and efficiency. Their most cited work, "FLTRNN: Faithful Long-Horizon Task Planning for Robotics with Large Language Models" (2024, 26 citations), introduces a novel framework that enhances planning fidelity by integrating LLMs with recurrent neural networks, addressing the limitations of in-context learning for extended tasks. Another key paper, "MLDT: Multi-Level Decomposition for Complex Long-Horizon Robotic Task Planning with Open-Source Large Language Model" (2024, 17 citations combined), proposes a hierarchical approach that breaks down intricate tasks into manageable sub-plans, leveraging open-source LLMs to democratize access to advanced planning capabilities. With a total of 43 citations across their top papers, Zhang’s work is gaining traction for its practical impact on robotics, offering scalable solutions that reduce dependency on proprietary models. Their achievements highlight a commitment to making robotic planning more faithful, efficient, and accessible, positioning them as a notable contributor to the intersection of AI and robotics.