About me: Hello world! You can call me Quan. I am an undergrad at HUST, Vietnam, and also an AI Resident at Qualcomm AI Research.

I like to build and poke at large language models, then ask why they work. I am especially interested in the theory behind LLM abilities: what principles explain the behaviors we observe, and how we can make them more reliable.

Outside of research, I love traveling to collect new experiences (and too many photos 🎞️ 🖼️) as my little way to “save” memories

Updates

• [Dec 2025] I am actively looking for a PhD position. Good luck to me!

• [Jun 2025] I will attend ICML 2025. Let’s connect!

• [May 2025] I got 2 papers accepted at ICML 2025.

• [Apr 2025] I officialy joined Qualcomm AI Research as an AI Resident.

• [Dec 2024] I achieved an IELTS Academic band score of 7.0!

• [Aug 2024] I joined VinAI Research as an AI Resident.

• [Apr 2024] My first paper was accepted at ICASSP 2024.

Research Interests

My long-term goal is to align the growing capabilities of modern AI with scientific transparency and universal access, building systems that are as interpretable as they are useful, including in resource-constrained settings.

I work on reliability and efficient reasoning for large language models, with two recurring themes: making reliability an explicit design objective (not just a byproduct of scaling), and turning theoretical principles into practical inference-time algorithms. Concretely, I explore robust fine-tuning and test-time compute, using Optimal Transport for structure-aware semantics and Sequential Monte Carlo as a principled framework to control diversity and uncertainty at inference time.

During my PhD, I hope to formalize these connections and translate them into simple, deployable methods for robust and efficient AI agents.