Nguyen Manh Tien Anh

I am currently an Associate Research Fellow at Deakin Applied Artificial Intelligence Initiative, where I research memory in Agentic AI and Multi-Agent Systems. My advisors in my current work are Dr. Hung Le and Prof. Svetha.

In 2023, I received my Engineer's degree in Information Technology from the School of Information and Communication Technology, Hanoi University of Science and Technology. My thesis, titled “Deep Learning in NLP for Anomalous HTTP Requests Detection,” explored the intersection of NLP and cybersecurity and was awarded as the Best Thesis/Presentation in the cybersecurity committee.

Previously, I was Head of Multimodal Foundation Models at the Institute of Automotive AI Development, VinFast. In this role, I led the development of foundation models for multimodal understanding and downstream applications such as virtual assistants and robotics. I also worked as a Research Assistant at the LISSI Tincnet laboratory, Université Paris-Est Créteil (UPEC), focusing on NLP applications for network management and cybersecurity.

Email  /  CV  /  GitHub  /  LinkedIn

Research Interests

My current research interests lie at Memory in Agentic AI systems, with a focus on how memory mechanisms can be designed to support reasoning, adaptation, and long-term interaction in intelligent agents. Inspired by neuroscience and human memory processes, I am interested in building agent architectures that can store, retrieve, and update knowledge over time in a robust and context-aware manner.

Previously, I also worked on NLP, Speech Technologies, and Cybersecurity, with applications spanning intelligent assistants, multimodal learning, and secure AI systems.

News

●   [Aug. 2026]   My paper, titled "Creative Generation via Multi-Agent Debate: Does Debate Suppress Diversity?," was accepted at EMNLP 2026 (Main-Oral).

●   [Feb. 2026]   I joined the Deakin Applied Artificial Intelligence Initiative as an Associate Research Fellow.

●   [Nov. 2025]   I was promoted to Head of Multimodal Foundation Models.

●   [Aug. 2025]   I was invited to serve as a reviewer for the IEEE Internet of Things Journal (Q1-ranked journal), in recognition of my expertise in AI and network systems.

●   [Feb. 2025]   My paper, titled "Zero Trust: Deep Learning and NLP for HTTP Anomaly Detection in IDS," has been accepted for publication in the IEEE Journal on Selected Areas in Communications (Q1-ranked journal, Impact Factor: 13.8).

●   [July 2024]   My proposed project, "AI Tutor for Kids," was selected for investment by Vintechtalent. We received $10K in funding for development, and I served as the project leader.

●   [June 2024]   My paper, titled "Improving Speech Recognition with Prompt-Based Contextualized ASR and LLM-Based Re-Predictor," has been accepted for publication at the 25th Interspeech (A-ranked conference).

●   [Apr. 2024]   I joined the Vintechtalent Club, which gathers young talents passionate about technology within Vingroup to develop breakthrough technology products for the Vietnamese community.

●   [Aug. 2023]   I successfully defended my thesis with the highest distinction. My thesis was awarded as the best thesis/presentation in the cybersecurity committee.

●   [July 2023]   My paper, titled "Deep Learning in NLP for Anomalous HTTP Request Detection," has been accepted for publication at the 2023 19th International Conference on Network and Service Management (CNSM).

●   [May 2023]   I was selected for the position of Principal AI Engineer after several months of internship at the Vingroup Big Data Institute, where my work focused on Automatic Speech Recognition and Natural Language Processing.

●   [Oct. 2022]   I was selected to join the Speech Processing Department at the Vingroup Big Data Institute as an AI Intern. My main research areas were Automatic Speech Recognition, Language Modeling, and Domain Adaptation.

●   [July 2022]   I was selected to join the Vingroup AI Engineer Training Program and ranked in the top 5 of the entrance selection.

●   [Dec. 2021]   My first paper, titled "When NLP Meets SDN: An Application to Global Internet Exchange Networks," was accepted for publication at the IEEE International Conference on Communications (ICC 2022).

This paper presents a Zero Trust-based intrusion detection system that leverages NLP, Deep Learning, and Transfer Learning to accurately and efficiently detect both known and novel anomalous HTTP requests.

This paper proposes a novel architecture integrating Large Language Models (LLMs) and prompt mechanisms to significantly enhance Automatic Speech Recognition (ASR) accuracy in challenging conditions.

This paper proposes a novel model combining NLP, DL, and the BERT model to differentiate normal from anomalous HTTP requests, achieving high accuracy on known threats and demonstrating transfer learning capability for unseen attack patterns.

This paper proposes an NLP-driven framework using a seq2seq LSTM model to enable natural language-based network configuration for Intent-Based Networking (IBN).

Awards and Honours

●   [2025]   Received a full PhD scholarship offer from Monash University (Declined).

●   [2025]   Received a full PhD scholarship offer from Paris-East Créteil University (Declined).

●   [2024]   Awarded $10K in funding for my AI Project in the VinTechTalent Program - Vingroup.

●   [2023]   Best Thesis/Presentation Award from the Cybersecurity Committee at the School of Information and Communication Technology.

●   [2022]   Excellent in Teaching Assistant Award in the School of Information and Communication Technology.

●   [2016-2018]   2nd and 4th Prize in the City-level Mathematics Competition for High School.

Thanks Jon Barron for the awesome website template!