University of Southampton representative ECR Board
Dr. Shuang Ao is a Research Fellow in AI for Good at the Responsible AI (RAi-UK), University of Southampton. Her research focuses on building safe and trustworthy large language models (LLMs) and responsible LLM-based agent systems, with emphasis on uncertainty quantification, automatic failure detection, and safety alignment for multimodal and safety-critical applications. Her PhD at The Open University (Knowledge Media Institute) explored uncertainty calibration, contrastive semantic similarity, and empirical optimal risk to enhance model trustworthiness. She has published in leading venues including UAI, EMNLP, and TACL, and received several awards such as the UAI Student Scholarship (2023–2024) and the Melete Award for PhD Students. Dr. Ao leads the AAAI 2026 Workshop on LLM-based Multi-Agent Systems (LaMAS), serves as Review Chair for UKAIRS 2025, and is advancing the development of safe, agentic LLM systems with autonomous reasoning and self-verification for responsible real-world deployment.