About

My name is Van Dai Do. I build efficient & safe LLMs—reinforcement learning, activation steering with memory, and retrieval-guided decoding. Additionally, I work on time-series forecasting, mostly with foundational models and edit-based methods. I am currently an Associate Research Fellow at Deakin University’s Applied AI Institute (A2I2).

News

💰 I received a grant from Lambda's Research Grant Program
🎉 1 paper accepted to NeurIPS 2026, and another accepted in TMLR
🔊 Our tutorial got accepted at ACML 2026! See you in Melbourne!
🎉 1 paper accepted to EMNLP 2026
🔊 I was awarded "Gold Reviewer" for ICML 2026!
🎉 1 paper accepted to ICML 2026
🎉 1 paper accepted to ACL 2026
📖 I submitted my PhD thesis titled “Efficient and Safe Large Language Models with Reinforcement Learning” after 2 years and 5 months on this journey! Very grateful for the support of my great supervisors Dr. Hung Le and Professor Svetha Venkatesh.
🔊 I am invited to present my ACL 2025 poster at Kingston AI Group Symposium 2026, held by Australian Institute for Machine Learning (AIML) at Adelaide!
🎉 1 paper accepted to TMLR with Journal to Conference (J2C) Certification!
🎉 1 paper accepted to ICDM 2025
🎉 I won the 3-Minute Thesis (3MT) competition hosted by Deakin’s A2I2. What a great journey—grateful for the experience and everyone who supported me along the way.
🎉 1 paper accepted to EMNLP 2025
🎉 1 paper accepted to ACL 2025
📖 I started working as a Research Assistant under Dr. Hung Le's supervision on the Australian Research Council (ARC) DECRA grant to implement innovative time series forecasting models.
🎉 1 paper accepted to ECAI 2024
📖 Invited tutorial at AAMAS
🛫 I started my PhD journey at Applied Artificial Intelligence Initiative (A2I2), Deakin University, Australia with a full-ride HDR scholarship!

Visitors

Live view of where visitors are coming from.

Publications

NeurIPS 2026
SteerCast: Retrieval-Based Latent Steering for Decoder-Only Time Series Forecasting
Van Dai Do, Huu Hiep Nguyen, Minh Hoang Nguyen, Hung Le
TMLR
CausalPlan: Empowering Efficient LLM Multi-Agent Coordination Through Causality-Driven Planning
Minh Hoang Nguyen, Van Dai Do, Dung Nguyen, Thin Nguyen, Hung Le
EMNLP 2026 (Oral)
Creative Generation via Multi-Agent Debate: Does Debate Suppress Diversity?
Tien Anh Nguyen, Khanh-Binh Nguyen, Van Dai Do, Svetha Venkatesh, Hung Le — arXiv
ACL 2026
SPaCe: Unlocking Sample-Efficient Large Language Models Training With Self-Pace Curriculum Learning
Van Dai Do, Manh Nguyen, Svetha Venkatesh, Hung Le — PDF
ICML 2026
Reviving Error Correction in Modern Deep Time-Series Forecasting
Minh Hoang Nguyen, Van Dai Do, Huu Hiep Nguyen, Dung Nguyen, Kien Do, Hung Le — OpenReview
ACL 2025
Dynamic Steering With Episodic Memory For Large Language Models
Van Dai Do, Quan Tran, Svetha Venkatesh, Hung Le — PDF · Code
EMNLP 2025
Sample Efficient Alignment Learning With Episodic Control
Van Dai Do, Quan Tran, Ahmed Kirmani, Lu Zhang, Hung Le
TMLR 2025
Reasoning Under 1 Billion: Memory-Augmented Reinforcement Learning for Large Language Models
Hung Le, Van Dai Do, Dung Nguyen, Svetha Venkatesh
ICDM 2025
Accelerating Long-Term Molecular Dynamics with Physics-Informed Time-Series Forecasting
Hung Le, Sherif Abbas, Minh Hoang Nguyen, Van Dai Do, Huu Hiep Nguyen, Dung Nguyen
ECAI 2024 (Oral)
Large Language Models Prompting With Episodic Memory
Van Dai Do, Quan Tran, Svetha Venkatesh, Hung Le — Code
IEEE TVT 2023
Deep RL for Multi-Hop Offloading in UAV-Assisted Edge Computing
Nguyen Tien Hoa, Van Dai Do, Le Hoang Lan, Nguyen Cong Luong, Duc Van Le, Dusit Niyato — Link

Grants

Lambda
Lambda Research Grant Program
GPU cloud compute credits for AI research

Professional Service

Reviewer — AAAI
Reviewer — TMLR
Reviewer — NeurIPS 2026
Reviewer — ICML 2026 Gold Reviewer
Reviewer — ACL Rolling Review (ARR)

Contact

Email: v.do@deakin.edu.au · Phone: +61 412 242 886