Siteng Huang

Papers

Total Citations

H-Index

About

Siteng Huang is an emerging researcher at the forefront of robotics and multimodal artificial intelligence, with a particular focus on vision-language-action (VLA) models for quadruped robot systems. His work bridges the gap between large language models and real-world robotic deployment, tackling some of the most pressing challenges in embodied AI. Huang's most notable contribution, "QUAR-VLA: Vision-Language-Action Model for Quadruped Robots" (2024), has already garnered 20 citations since its publication, demonstrating rapid uptake within the robotics and AI communities. This work establishes a foundational framework for enabling quadruped robots to interpret and act upon complex multimodal instructions — a significant step toward truly intelligent autonomous systems. Building on this foundation, his 2025 follow-up work, "Quart-Online," directly confronts the practical challenge of inference latency when deploying multimodal large language models in real-time robotic contexts, pushing beyond conventional parameter reduction techniques to preserve model performance without sacrificing responsiveness. Together, these contributions mark Huang as a researcher deeply committed not just to theoretical innovation, but to bridging cutting-edge AI research with deployable, real-world robotic solutions — making his work essential reading for anyone exploring the intersection of language models and autonomous robotics.