Jing HAO

I'm pursuing my Ph.D. degree in the University of HongKong, Faculty of Dentistry (Ranking 2nd in the world), specializing in Medical AI and MLLM, supervised by Prof. Kuo Feng Hung and Prof. Tsoi, James Kit Hon.

Previously, I worked as a Computer Vision Engineer on Baidu VIS from 2022.07 to 2024.08. I received my M.S. degree in Huazhong University of Science and Technology (HUST, 2022), and B.S. degree in Chinese University of Mining and Technology (CUMT, 2020).

My research interests span the area of computer vision, self-supervised pre-training, multimodal large language model (mllm), and AI4Science.

e-mail  /  Scholar  /  github  /  Wechat  /  ORCiD

Publications

See full list at Google Scholar. (* indicates equal contribution, # indicates corresponding author)

We present OralAgent, the first dental-specialized AI agent that unifies multimodal reasoning, tool-based decision-making, and knowledge-grounded retrieval within an end-to-end automated framework.

We present OralGPT-Omni, the first dental-specialized MLLM designed for comprehensive and trustworthy analysis across diverse dental imaging modalities and clinical tasks. We also introduce MMOral-Uni, the first unified multimodal benchmark for dental image analysis.

We present OralGPT-Plus, an agentic vision–language model designed to perform iterative and symmetry-aware diagnostic reasoning for panoramic dental radiograph analysis.

We introduce MMOral, the first large-scale multimodal instruction dataset and benchmark tailored for panoramic X-ray interpretation. We also propose OralGPT, a multimodal vision-language model for panoramic X-ray analysis.

Open-source oral-maxillofacial imaging datasets were identified through electronic databases and dataset platforms. 105 datasets with 437538 images and 100 intraoral videos from patients across twenty-one countries were included.

T-Mamba is the first work to introduce frequency-based features into vision mamba, its flexibility allows it to process both 2D and 3D tooth data without the need for separate modules.

The first to propose exploring to the solution of glass surface segmentation by fully harnessing the capabilities of existing VFMs.

A strong framework for utilizing Multiple datasets to pretrain DETR-like detectors without the need for manual label spaces integration.

We participated in the challenge of “MICCAI STS 2024: Panoramic X-ray Images”, and ranked 6th among all submitted teams.

We designed a FullAnno system, which is a data engine that can generate large-scale, high-quality, and fine-grained image caption datasets automatically.

This study proposed a novel semi-supervised transformer-based framework designed for automated tooth segmentation and identification on panoramic radiographs.

A self-attention approximation without training parameters which captures global spatial features with linear complexity.

A stronger stitching algorithm for fisheye images by combining the traditional image processing method with deep learning.

A signs recognition framework based on convolutional neural networks (CNNs) for weld images.

Education & Experience

The University of Hong Kong (HKU) 2024.09 - now Ph.D Student, Faculty of Dentistry (Ranking 3rd in the world) MLLM, Agent

Ant Group 2026.05 - now Visiting Student (Ant Star⭐️)

Shanghai AI Lab 2026.01 - 2026.05 Visiting Student

Baidu VIS 2022.07 - 2024.08 Computer Vision Engineer

Huazhong University of Science and Technology (HUST) 2020.09 - 2022.06 Master Student, School of Mechanical Science and Engineering

China University of Mining and Technology (CUMT) 2016.09 - 2020.06 Undergraduate Student, School of Mechanical and Electrical Engineering