Sharan Maiya

CV

📄 Download PDF

Education

University of Cambridge

PhD Applied Linguistics, Oct 2024 – 2027
MRes AI for Environmental Risks, Oct 2023 – 2024

Imperial College London

MSc Statistics, Oct 2020 – Sep 2021

The University of Edinburgh

BSc Computer Science and Mathematics, Sep 2016 – Jun 2020

Research & Work

ML Alignment and Theory Scholars, Scholar
Jan 2025 – Present
Cadenza Labs, Co-Founder and Research Engineer
Aug 2024 – Present
Supervised Program for Alignment Research, Student Researcher
Oct 2023 – Jun 2024
Cambridge AI Safety Hub, Fellowship Facilitator
Winter 2023 and Winter 2024
University of Edinburgh, Pre-Doctoral Research Assistant
Sep 2021 – Jun 2023

Papers

Sharan Maiya, Henning Bartsch, Nathan Lambert, and Evan Hubinger. "Open Character Training: Shaping the Persona of AI Assistants through Constitutional AI". arXiv, 2025. Link
Sharan Maiya, Yinhong Liu, Ramit Debnath, and Anna Korhonen. "Improving Preference Extraction in LLMs By Identifying Latent Knowledge Through Classifying Probes". ACL, 2025. Link
Walter Laurito*, Sharan Maiya*, Gregoire Dhimoila*, Owen Ho Wan Yeung, and Kaarel Hanni. "Cluster-Norm for Unsupervised Probing of Knowledge". EMNLP, 2024. Link