Jiayang Song
Papers
9
Total Citations
118
H-Index
5
About
Jiayang Song is a researcher at the forefront of artificial intelligence, robotics, and cyber-physical systems (CPS), with a particular focus on integrating large language models (LLMs) into autonomous robotic planning and safety-critical applications. His most influential work, "ISR-LLM: Iterative Self-Refined Large Language Model for Long-Horizon Sequential Task Planning" (2024), has garnered 56 citations and introduced a novel framework enabling robots to autonomously refine task plans across complex, extended sequences — a significant advance in making LLM-powered robotics more reliable and adaptive. Complementing this, his contributions to AI-enabled CPS, including industrial benchmarks built on NVIDIA Isaac Sim and the semantics-guided safety enhancement framework, demonstrate a commitment to bridging cutting-edge AI research with real-world deployment concerns. Song has also pioneered testing methodologies for emerging Vision-Language-Action (VLA) models through works like VLATest and LADEV, addressing the critical but often overlooked challenge of evaluating multimodal robotic systems. With over 100 cumulative citations across his growing body of work, Song is establishing himself as a key voice in ensuring that next-generation AI-driven robotic systems are not only capable, but rigorously validated and safe for industrial and everyday use.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9