Xingyao Wang

Shandong University of Science and Technology

Papers

Total Citations

H-Index

About

Xingyao Wang is a rising researcher at the forefront of artificial intelligence and marine robotics, with key contributions spanning underwater computer vision and large language model (LLM) agent architectures. In underwater object detection, Wang developed an innovative algorithm integrating channel and spatial fusion attention mechanisms, directly addressing the challenges of poor detection precision in complex, low-visibility marine environments—a foundational technology for autonomous underwater vehicle operations. This work has garnered 20 citations, establishing Wang’s early impact in marine perception systems. More recently, Wang has advanced the field of LLM agents, proposing a paradigm-shifting approach that replaces traditional JSON or text-based action generation with executable code actions. This method, detailed in a 2024 paper with 9 citations, significantly enhances agent robustness and flexibility by leveraging code’s structured, verifiable nature, enabling more reliable tool use and robot control. Wang’s dual focus on practical, high-stakes applications—from deep-sea exploration to autonomous decision-making—demonstrates a unique ability to bridge perception and action. As an emerging voice in AI-driven robotics, Wang’s work promises to reshape how machines interact with both the physical world and complex software environments.