Scholar
Yining She
Google Scholar ID: UdXrcOoAAAAJ
PhD Student in Software Engineering, Carnegie Mellon University
Citations & Impact
Citations
13
H-index
2
i10-index
1
Publications
3
Co-authors
6
list available
Publications
Resume (English only)
Academic Achievements
- Selected publications include 'RAG Makes Guardrails Unsafe? Investigating Robustness of Guardrails under RAG-style Contexts' (Under Review, 2025), which investigates how robust LLM-based guardrails are against additional information embedded in the context, and 'FairSense: Long-Term Fairness Analysis of ML-Enabled Systems' (to appear in 2025 IEEE/ACM 47th International Conference on Software Engineering), proposing a simulation-based framework to detect and analyze long-term unfairness in ML-enabled systems.
Research Experience
- Recent work focuses on safeguarding Large Language Model Agents by identifying provenance of agent action to systematically prevent risky actions caused by underspecified user intent and LLM hallucination. Previously developed FairSense for proactive analysis of long-term fairness issues that specifically considers feedback loop interaction between AI system and environment.
Education
- Currently a fourth-year PhD student in Software Engineering at the Software and Societal Systems Department, Carnegie Mellon University, advised by Dr. Eunsuk Kang. Received a Bachelor of Engineering in Computer Science from ShanghaiTech University, where he worked with Dr. Zhihao Jiang at the Human-Cyber-Physical Systems Lab.
Background
- Research interests lie at the intersection of Software Engineering (SE) and Artificial Intelligence. Designs methods and builds tools to empower developers to evaluate and analyze the safety, robustness, & fairness of AI systems, and design better systems.
Miscellany
- Also a part-time researcher collaborating with OCI GenAI.
