Vo Hoang Nhat Khang (Chris) - PhD Student in NLP

Vo Hoang Nhat Khang (Chris)

PhD Student in Natural Language Processing

PhD · MBZUAI

I am a PhD student in the Computing & Mathematical Sciences Division, advised by Professor Timothy Baldwin and Dr. Yova Kementchedjhieva.

Teaching Assistant · MBZUAI

I am currently a teaching assistant for Mathematical Foundations for AI, co-organizing the lab sessions and creating problems for the final examination.

Background. I completed my Computer Science degree at HCMUT-VNU in an accelerated 3.5 years. Under the guidance of Assoc. Prof. Dr. Tho T. Quan, I built foundations in machine learning, deep learning, and multimodal analysis, with programming experience in Python, C++, and Java.

Research interests. Vision-language models, uncertainty quantification, hallucination detection, and multimodal learning.

News

A running log of research updates and recent milestones.

Saigon Notre-Dame Cathedral Basilica in Ho Chi Minh City, featured by CSoNet 2026 Ben Thanh Market in Ho Chi Minh City, featured by CSoNet 2026

Paper Accepted to CSoNet 2026

Entropy-Weighted Neighborhood Gradients for Decentralized Federated Learning with Noisy Medical Labels was accepted to CSoNet 2026 and will appear in Springer's Lecture Notes in Computer Science (LNCS, Q2). Congrats everyone!

Conference website

Mohamed bin Zayed University of Artificial Intelligence logo

Teaching Assistant for Mathematical Foundations for AI

I am currently a teaching assistant for Mathematical Foundations for AI at MBZUAI, where I co-organize the lab sessions and create problems for the final examination. I will release the lab materials and slides I prepared for students.

Materials coming soon

Example MERIT repair episode: error classification, hybrid retrieval of a positive precedent, LLM repair generation, and validation

Paper Accepted to REALM @ EMNLP 2026

Causal Episodic Memory for Feedback-Driven Agent Repair was accepted to REALM, the Second Workshop for Research on Agent Language Models, co-located with EMNLP 2026 in Budapest, Hungary.

PaperCodeWorkshop

SHROOM-Visions shared task logo

Solo-Author Paper Conditionally Accepted at an EMNLP Workshop

Team HalluciNauts at SHROOM-Visions: Anchor-Preserving Dual-Resolution Grounding for Fine-Grained Vision Hallucination Detection was conditionally accepted to SHROOM-Visions at the Uncertainty in NLP workshop, co-located with EMNLP 2026. Proceedings or arXiv link coming soon.

Shared task

Academic Journey

Khang with Professor Timothy Baldwin at MBZUAI Khang with Dr. Yova Kementchedjhieva at MBZUAI

PhD Student at MBZUAI

Working under the guidance of Professor Timothy Baldwin and Dr. Yova Kementchedjhieva, focusing on uncertainty in multimodal learning and vision-language models.

Khang at the National University of Singapore campus Khang having lunch with colleagues at NUS

Visiting Scholar at NUS

Worked at the National University of Singapore (NUS) with Professor Anji Liu on uncertainty quantification and hallucination detection in vision-language models.

Ho Chi Minh City skyline featured by CSoNet 2026

Program Committee Member for CSoNet 2026

Served as a reviewer and Program Committee member for the 15th International Conference on Computational Science and Network Intelligence (CSoNet 2026).

Conference website

Selected Research Publications

A few papers that escaped my laptop and made it into the world.

MERIT framework using error-typed retrieval and dual-polarity episodic memory for iterative Text-to-SQL repair
2026

Causal Episodic Memory for Feedback-Driven Agent Repair

Khang Nhat Hoang Vo, Tam Minh Chu, Anh Trac Duc Dinh, Thuyen Vinh Ha Bui, Tho Quan

REALM Workshop @ EMNLP 2026 · 2026

Paper GitHub
Comparison of constrained image patch prediction and ambiguous linguistic alternatives in JEPA models
2026

The JEPA Paradox in Language: The Geometry of Linguistic Alternatives

Anh Trac Duc Dinh*, Khang Nhat Hoang Vo*

* Equal contribution.

arXiv · 2026

Paper
Word clouds showing frequent tokens in the Cham and Khmer corpora Map showing Tay, Nung, Khmer, and Cham populations across Vietnam
2026

Echoes Across Vietnam's Highlands, Delta, and Coast: A Multilingual Corpus for Cham, Khmer, and Tay-Nung

Anh Trac Duc Dinh*, Khang Nhat Hoang Vo*, Vinh Cong Doan, Tai Tien Ta, Khoa Duc Anh Lam

* Equal contribution.

arXiv · 2026

Paper
Examples distinguishing visual recognition failures from knowledge failures in vision-language models Flowchart attributing vision-language model errors to visual evidence, unknown entities, unknown facts, or unrecallable facts
2026

Failing to See or Failing to Know? Attributing Errors in Vision-Language Models

Khang Nhat Hoang Vo, Artem Vazhentsev, Artem Shelmanov, Timothy Baldwin, Yova Kementchedjhieva

arXiv · 2026

Paper PDF
HuTieuBERT architecture with adaptive boundary-token fusion and morpheme-aware attention bias
2026

When Morphology Hides in Plain Sight: Breaking the Isolation in Vietnamese and Beyond

Anh Trac Duc Dinh, Khang Hoang Nhat Vo, Tai Tien Ta, Vinh Cong Doan, Tho Quan

ACL 2026 Main Conference · 2026

Paper Project page GitHub
Two-day AI Agent Hackathon agenda mapped to learning outcomes
2026

Catching the First Light of Tomorrow: A Hackathon-Based Framework for Introducing High School Students to AI Agents

Long Nguyen, Duc Nguyen, Quan Bui, Dung Phan, Dung Le, Khanh Nguyen, Quynh Vo, Khang Vo, Nam Duong, Anh Dinh, Tri Trinh, Chi Phan, An Nguyen, Thai Nguyen, Dang Le, Vinh Dang, Tho Quan

AAAI Press · Feb 1, 2026

Paper
Contrastive learning architecture aligning lumbar MRI images with diagnostic reports
2025

Revolutionizing Precise Low Back Pain Diagnosis via Contrastive Learning

Thanh Binh Le*, Hoang Nhat Khang Vo*, Tan-Ha Mai, Trong Nhan Phan

* Equal contribution.

Springer · Dec 31, 2025

Paper
TI-JEPA architecture using text-image cross-attention to predict masked image regions
2024

TI-JEPA: An Innovative Energy-based Joint Embedding Strategy for Text-Image Multimodal Systems

Khang H. N. Vo*, Duc P. T. Nguyen*, Thong Nguyen, Tho T. Quan

* Equal contribution.

SOICT

Paper GitHub
Vietnamese-Bahnar translation pipeline with segmentation and mapping phases
2024

Revitalizing Bahnaric Language through Neural Machine Translation: Challenges, Strategies, and Promising Outcomes

Hoang Nhat Khang Vo, Duc Dong Le, Tran Minh Dat Phan, Tan Sang Nguyen, Quoc Nguyen Pham, Ngoc Oanh Tran, Quang Duc Nguyen, Tran Minh Hieu Vo, Tho Quan

AAAI Press

Paper

Featured Projects

Things I built after saying, "This should be a quick experiment".

2025 - VitalFit Persona: AI-powered healthcare application with agentic AI capabilities. GitHub

2025 - V7: advanced AI system showcasing machine learning and AI implementations. GitHub

2025 - Synthetic Generator: data generation system for synthetic datasets for machine learning. GitHub

2025 - Namizer: Vietnamese tokenizer that breaks words into 5 components. GitHub

2024 - HCMUT Chatbot: LLM and RAG-based chatbot for university academic admission. GitHub

Media Coverage & Recognition

Thanh Nien Newspaper - Early graduation with near-perfect thesis score. Article

Znews.vn - Young engineer uses AI to change Vietnamese typing. Article

OISP HCMUT - Bach Khoa International student shines in AmCham Scholarship 2024. Article

OISP HCMUT International - From barely in to top talented student. Article

OISP HCMUT - Rejecting New Zealand study abroad to choose Bach Khoa. Article

BK Quoc Te Facebook - Bach Khoa International Program feature. Video

Thanh Nien Facebook - Social media feature on academic achievements and early graduation. Post

BK Quoc Te Facebook - Academic excellence recognition. Post

Get In Touch

Want to discuss research, build something useful, or compare notes on why a model is confidently wrong? Say hello.

Email - nhatkhangdtp [at] gmail [dot] com · Khang.Vo [at] mbzuai [dot] ac [dot] ae

Location - Abu Dhabi, UAE (PhD studies) · Ho Chi Minh City, Vietnam

Links - LinkedIn · GitHub · Google Scholar