Introduction

👋 My name is Quang-Minh (I go by Minh too). I am a researcher at KAIST. I previously obtained my master’s degree in Data Science from KAIST Graduate School of Data Science, where I worked with Professor Taegyoon Kim, and my bachelor’s degree from KAIST with a double major in Electrical Engineering ⚡ and Mathematical Sciences 🧮.

I am curious about AI interpretability and safety 🤖. This encompasses many big questions, both behavioural and representational, from how models handle fundamental concepts (truth, stance, difficulty, beliefs, etc.) to high-stakes applications involving social-political threats.

If you are interested in collaboration, please don’t hesitate to reach out. Feel free to email me, or we can grab coffee if you pop by Daejeon!

Selected Papers

Quang Minh Nguyen and Luis Frentzen Salim. “Whether LLMs Can Navigate Beliefs and Facts Depends on How You Phrase It” (EMNLP 2026). [paper] [code]

Quang Minh Nguyen, Uzair Ahmed, and Taegyoon Kim. “Can LLMs Reliably Self-Report Adversarial Prefills, and How?” (EMNLP 2026). [paper] [code]

See all publications.

News

📌 October 2026: I’ll be at EMNLP 2026 in Budapest presenting two papers. Come say hi!

• October 2026: I made a new Substack blog ✍️
• September 2026: Excited to be taking BlueDot Impact’s Technical AI Safety course!
• August 2026: 🎊 Three of our papers have been accepted to EMNLP 2026: “Whether LLMs Can Navigate Beliefs and Facts Depends on How You Phrase It” and “Can LLMs Reliably Self-Report Adversarial Prefills, and How?” to the Main Conference, and “Towards Reliable Truth-Aligned Uncertainty Estimation in Large Language Models” to Findings! 🎊

Older news

• August 2026: Our new preprint “Whether LLMs Can Navigate Beliefs and Facts Depends on How You Phrase It” has been released!
• June 2026: Our new preprint “Can LLMs Reliably Self-Report Adversarial Prefills, and How?” has been released!
• June 2026: Our ongoing work, “Can LLMs Navigate Beliefs and Facts? Depends on How You Phrase It”, will be presented at the Philosophy Meets Machine Learning Workshop at ICML 2026!
• May 2026: Our paper “The Broken Telephone Changes Tone: Examining Nuanced Linguistic Cues in LLM Chains-of-Translation” has been accepted to the Workshop on Multilinguality in the Era of LLMs at ACL 2026! It will also be presented at the Trustworthy AI for Good Workshop at ICML 2026. Connect with me in Seoul!
• February 2026: Our paper “Analysing LLM Persona Generation and Fairness Interpretation in Polarised Geopolitical Contexts” has been accepted to the Student Research Workshop at EACL 2026!
• September 2025: I will present our ongoing work, “Investigating Difficulty Perception in the Reasoning of LLMs”, at the MathNLP Workshop at EMNLP 2025! Let’s connect in Suzhou.
• May 2025: Our paper “Is External Information Useful for Stance Detection with LLMs?” has been accepted to the 🎊Findings of ACL 2025🎊! I’ll be in Vienna to present; come say hi if you’ll be there too!
• August 2024: I graduated from the undergraduate program at KAIST and will continue my journey here as a master’s student.

Miscellaneous

I enjoy learning about history 📜, especially East Asian history, as well as playing musical instruments 🎵 (I play the piano, guitar, ukulele, and harmonica). My favourite game is Total War: Three Kingdoms 🎮.