cv

Master of Science in Computer Vision

• Current GPA: 3.93/4 (old scale, 24 credits), 4.0/4 (US scale).
• Relevant courses: Visual Object Recognition and Detection; Medical Imaging: Physics and Analysis.

Bachelor of Science in Data Science and Artificial Intelligence

• GPA: 3.96/4 (126 credits). Ranked 2nd out of 108 students in the major and 8th out of 7,000 students in Cohort 65.
• Thesis: Mamba-based Contrastive Language-Image Pretraining for Endoscopic Image Analysis (9/10).
• Relevant courses: Machine Learning, Introduction to Deep Learning, Computer Vision, Natural Language Processing.

MedEmoBench: Revealing and Mitigating Emotion-Induced Inconsistency in Medical Vision-Language Models

• Minh Khoi Ho, Omar Abdelshafi Mohamed, Xinyue Bi, Junyuan Hong, Zhipeng Deng, Yefeng Zheng, Yutong Xie.
• Preprint.

Attention-Based Prototype Calibration for Multi-Rater Few-Shot Medical Image Segmentation

• Truong Vu, Minh Khoi Ho, Yutong Xie.
• MICCAI 2026.

Can Induced Emotion Bias LLM Behaviors in Sequential Decision Making?

• Minh Khoi Ho, Zihao Zhu, Runchuan Zhu, Levina Li, Zhiwen Fan, Zhangyang Wang, Junyuan Hong.
• COLM 2026.

Beyond the Global Scores: Fine-Grained Token Grounding as a Robust Detector of LVLM Hallucinations

• Dung Nguyen, Minh Khoi Ho, Qi Chen, Yutong Xie, Nguyen Cam-Tu, Minh Khoi Nguyen, Dang Huy Pham Nguyen, Anton van den Hengel, Johan W Verjans, Phi Le Nguyen, Vu Minh Hieu Phan
• CVPR 2026.

Overthinking Causes Hallucination: Tracing Confounder Propagation in Vision Language Models

• Abin Shoby, Ta Duc Huy, Tuan Dung Nguyen, Minh Khoi Ho, Qi Chen, Anton van den Hengel, Phi Le Nguyen, Johan W. Verjans, Vu Minh Hieu Phan.
• CVPR 2026 Findings.

Localizing Before Answering: Segmentation-Guided Large Multi-modal Model for Grounded Medical Visual Question Answering

• Dung Nguyen*, Minh Khoi Ho*, Huy Ta, Thanh Tam Nguyen, Qi Chen, Kumar Rav, Quy Duong Dang, Satwik Ramchandre, Son Lam Phung, Zhibin Liao, Minh-Son To, Johan Verjans, Phi Le Nguyen, Vu Minh Hieu Phan
• IJCAI 2025. *Co-first authors.

TrafficVLM: A Controllable Visual Language Model for Traffic Video Captioning

• Quang Minh Dinh, Minh Khoi Ho, Anh Quan Dang, Hung Phong Tran.
• CVPR 2024 Workshop.

Visiting Researcher

• Advisor: Prof. Yefeng Zheng. Participated in Westlake's REU program.
• Project: MedEmoBench: Revealing and Mitigating Emotion-Induced Inconsistency in Medical Vision-Language Models.
• Built MedEmoBench, a framework covering 340k+ samples and 5 medical VLM tasks to evaluate how user-expressed emotions affect responses.
• Evaluated 4 hallucination mitigation methods and proposed a GRPO-based approach to reduce emotion-induced inconsistencies, which improved accuracy by 2% and consistency by 5%.

Visiting Researcher

• Advisor: Assistant Prof. Yanda Meng.
• Project: Agentic AI for ECG interpretation (first-authored work, in progress).
• Investigating agentic approaches to ECG interpretation, with a focus on continual learning from clinical guidance and user feedback.

Research Assistant

• Advisor: Assistant Prof. Yutong Xie.
• Current project: Agentic AI for ECG interpretation (joint work with KAUST).
• Past project (1): Safety and robustness of Medical Vision-Language Models (joint with Westlake University).
• Past project (2): Multi-rater image segmentation.
• Proposed the research topic, curated and preprocessed 3 multi-rater segmentation datasets, and benchmarked existing multi-rater and few-shot segmentation methods (published at MICCAI 2026).

Research Intern

• Advisor: Assistant Prof. Junyuan Hong.
• Investigated how 5 induced emotions bias LLM behavior in sequential decision-making tasks, including the Iowa Gambling Task and two-armed bandit (published at COLM 2026).

Research Assistant

• Advisor: Assoc. Prof. Nguyen Phi Le.
• Member of the Applied AI in Healthcare group, in collaboration with the University of Adelaide.
• Project: Grounding and explainability in vision-language models.
• Developed the Attention Dispersion Score (ADS) for hallucination detection, 1/2 model components that achieve up to 90% accuracy in token-level hallucination detection (published at CVPR 2026).
• Developed a model integrating LISA and PromptHighlighter to localize organs and reduce hallucinations in chest X-ray VQA, with 2-4% F1 improvement in MIMIC and VinDr VQA (published at IJCAI 2025).

Research Assistant

• Advisor: Dr. Dinh Viet Sang.
• Member of the Computer Vision research team.
• Conducted literature reviews, tested programming environments, and benchmarked methods for computer vision and medical imaging tasks, including facial keypoint detection, polyp segmentation, pose estimation, and multimodal learning.

Teaching Assistant

• Course instructor: Prof. Hung Son Nguyen.
• Introduction to Deep Learning, semester 2023.1.
• Prepared 5 homework sessions, 3 lab courses, and materials for approximately 100 students.
• Hosted 4 capstone project evaluation sessions for 16 groups.

Data & Software Engineer

• Integrated data from internal management systems such as Jira to evaluate employee performance using ETLPentaho and Apache NiFi, helping decrease workload for the HR department by 80%.
• Built a system to conduct internal surveys for 1500 employees and provide biannual rating reports automatically on Oracle APEX with modifications of JavaScript, CSS, and storage on Oracle SQL Server, reducing a 15 man-day task to 0.5 man-days.

AI Research Intern

• Researched segmentation-based models derived from Unet++ and DNA-Net; object tracking method based on SORT and ensemble two models to improve detection accuracy for the IRST (InfraRed Search and Track)System by 5% compared to non-deep learning methods.
• Implemented models using Python libraries: PyTorch, OpenCV; serving with ONNX and TensorRT.

• Travel Grant to MICCAI 2026
  • Awarded by MBZUAI. ~11000 AED award.

• DAAD Scholarship for STEM Disciplines in Germany (declined)
  • German Academic Exchange Service (DAAD).
• Padova Excellence Scholarship (declined)
  • University of Padua.
• Rising Scholars Fellowship Program (RSFP)
  • Singapore Management University. 20/120 applicants, 2000 SGD award

• HUST Certificate for Excellent Graduates
  • One of 140 recipients among 4,500 graduates in semester 2024.2.
• 3rd Place - AI City Challenge 2024, Track 2
  • AI City Challenge 2024.

• Exness Scholarship
  • One of 35 recipients among 250 applicants. 35M VND award.

• 6x HUST Excellent Scholarship
  • Scholarship for the top 5% of students: 2020 (1), 2021 (1), 2022 (2), 2023 (2).