Kureha Yamaguchi

Research Engineer @ The Alan Turing Institute | London, UK

prof_pic.jpg

Hi! I’m Kureha (pronounced cray.ha)- an AI Researcher at UK’s National Institute for Data Science and AI, also known as The Alan Turing Institute. I investigate AI system vulnerabilities and develop secure ML solutions, delivering impact-driven frontier research and engineering within the Turing’s Defence and National Security programme.

Prior to research, I was at the University of Cambridge completing my undergraduate and master’s degrees in Information and Computer Engineering, supported by a scholarship from the Institution of Engineering and Technology.

I love coding, and I really care about problem-solving in a socially responsible way that actually benefits humanity. Still figuring it all out and enjoying the process- thanks for visiting my page!

selected publications

  1. Where Do Reasoning Models Refuse?
    Kureha Yamaguchi, Benjamin Etheridge, and Andy Arditi
    ICML, Mechanistic Interpretability Workshop, 2026
  2. 2DSig-Detect: a semi-supervised framework for anomaly detection on image data using 2D-signatures
    Xinheng Xie, Kureha Yamaguchi, Margaux Leblanc, Simon Malzard, Varun Chhabra, Victoria Nockles, and Yue Wu
    Pattern Recognition Journal, Elsevier, 2025
  3. An AI red team playbook
    Anna Raney, Shiri Bendelac, Keith Manville, Mike Tan, and Kureha Yamaguchi
    Proceedings of SPIE, Jun 2024
  4. An AI blue team playbook
    Mike Tan, Kureha Yamaguchi, Anna Raney, Victoria Nockles, Margaux Leblanc, and Shiri Bendelac
    Proceedings of SPIE, Jun 2024