Paul He

prof_pic.jpg

I am a PhD student in Computer Science at Nanyang Technological University (NTU), Singapore, where I study reliability in large language models with a focus on methods that offer provable guarantees. Prior to this, I completed my BSc in Computer Science (Honours) at the University of Toronto and spent a year at ETH Zürich taking graduate-level courses in machine learning and NLP. I also worked as an Applied Scientist at AWS, where I conducted research on LLM reliability and evaluation in production systems.

You can view my CV here(Last Updated: 04.Jan.2026), and my google scholar here.

Email: hepaul [at] cs [dot] toronto [dot] edu

news

Jan 12, 2026 Starting my PhD at NTU Singapore!
Jan 04, 2026 Uncovering Hidden Correctness in LLM Causal Reasoning via Symbolic Verification has been accepted to EACL (main)! Preprint here.
Sep 01, 2025 Started as an Applied Scientist at AWS in Tuebingen!
Jun 19, 2025 Finished my undergraduate at UofT in computer science!
Jun 03, 2025 Our paper CCG Revisited: A Multilingual Empirical Study of the Kuhlmann-Satta Algorithm has been accepted to SIGPARSE: International Conference on Parsing Technologies. You can find the paper on the ACL Anthology here.
Jan 23, 2025 Our paper Pointwise Mutual Information as a Performance Gauge for Retrieval-Augmented Generation that I’ve contributed to can be found at here (arXiv) and has been accepted to NAACL 2025 (Main). My main contribution involves in section 4 of the paper.