About
Experience
Carnegie Mellon University
Postdoctoral Associate
Dec 2025 — Present
Multimodal Large Language Models and Embodied AI. Working on long-context diagnostic reasoning in clinical
LVLMs, 3D spatial reasoning over volumetric medical data (SpatialMed), and VLA benchmarking for biological
lab automation.
Seeing Machines (LSE-listed)
Machine Learning Scientist
Jun 2022 — Apr 2024
AI for Smart Vehicles. Led end-to-end development of driver hand activity recognition (3D hand-position
estimation MPJPE 53→32 mm; activity accuracy 37%→79%) and attention monitoring (87%→95%). Built a
diffusion-based photorealism enhancement pipeline (Diffusers, DreamBooth, LoRA) lifting hand-detection mAP
from 0.67 to 0.89. Shipped quantized inference models to production vehicle programs for Volkswagen, BMW,
and Ford.
Education
Australian National University
PhD, Computer Vision & Machine Learning
2018 — 2022
Peking University
MSc, Computer Science [Rank
3/46 (Top 5%)]
2013 — 2016
Sun Yat-sen University
BSc, Computer Science [Rank
3/252 (Top 5%)]
2009 — 2013
Technical Skills
LLM & MLLM
Post-training (SFT, LoRA, DPO, GRPO), LLaVA, InternVL, MedGemma
Generative Models
Diffusion Models, Stable Diffusion, ControlNet, DreamBooth
Foundations
CNNs, Transformers, PyTorch, Python
Academic Service
Reviewer
CVPR, ICCV, ECCV, ICML, NeurIPS, ICLR, and related venues
Supervisor
Associate PhD Supervisor at the Australian National University
Honors & Awards
ANU Merit International PhD Scholarship
2018 — 2022
ANU International PhD Scholarship
2018 — 2022
Postgraduate Research Scholarship (Data61–CSIRO)
2018 — 2022