Zhaoxia Yin
Papers
1
Total Citations
14
H-Index
1
About
Dr. Zhaoxia Yin is a leading researcher in embodied intelligence and adversarial machine learning, with a focus on the intersection of large language models (LLMs) and physical-world AI systems. Her work critically examines the security and robustness of decision-level processes in LLM-based embodied models, revealing how adversarial attacks can compromise agents that rely on language-guided planning. In her highly cited 2024 paper, "Exploring the Robustness of Decision-Level Through Adversarial Attacks on LLM-Based Embodied Models," she demonstrates that even sophisticated embodied agents—which integrate perception, language understanding, and real-world action—are vulnerable to targeted perturbations, underscoring the need for resilient architectures. This contribution has quickly garnered 14 citations, highlighting its timely impact on the AI safety community. Dr. Yin’s research bridges natural language processing and robotics, advancing our understanding of how to secure autonomous systems that operate in dynamic environments. Her work is essential for students and researchers developing trustworthy AI, as it provides foundational insights into the fragility of decision-making under adversarial conditions.
Research Focus
Key Achievements
Top Papers
- 1