KMi is thrilled to announce that Shuang Ao has successfully defended her PhD on 22 October 2024. Her thesis, titled “Building Trustworthy AI: Uncertainty Quantification and Failure Detection in Large Vision-Language Models,” addresses key challenges in making AI systems more reliable and trustworthy. Shuang’s research introduces new tools that help improve how well AI systems can assess their own accuracy and identify mistakes. These innovations are especially important for applications where safety is critical, as they help ensure AI systems are more dependable and better aligned with real-world needs, including natural language generation.

Under the supervision of Professors Advaith Siddharthan and Stefan Rueger, Shuang’s research offers important contributions to AI by helping align model confidence with accuracy and providing nuanced failure assessments.

Shuang has now embarked on her postdoctoral journey at the University of Southampton, where she is focused on building responsible large language models and vision-language models for healthcare. We wish her all the best in this exciting new chapter!