PhD Candidate · University of Liverpool
I develop accountable urban AI systems that connect foundation models, multimodal sensing, spatiotemporal prediction, infrastructure resilience, and planning decision support.
I develop accountable urban AI systems for public decision-making. My work explores transportation, urban climate, urban forestry, land-use planning, and human-centered interaction through a shared core: urban AI intelligence built on foundation models, multimodal sensing, spatiotemporal prediction, infrastructure resilience, and planning decision support.
My work centers on urban spatial intelligence: how urban actors behave across space and time. I develop frontier AI methods, including world models and geospatial foundation models, to reveal generalizable laws of urban science and build rigorous, theory-driven urban AI.
My research is organized around a central question: how can urban AI systems become auditable, interpretable, and useful for public institutions?
HGMM
MorphNet
UEA-MOE
DWTIF
Multi-Agent Recommendation System
ADAPT
Frequency as identity: A Fourier hypernetwork for spatiotemporal forecasting.
STPIA
Sustainable Project Governance: Scientometric Analysis and Emerging Trends.
Scientometric Governance Analysis
Campus Living Labs for Sustainability: How Multicultural University Communities Drive Nature-based Innovation.
Campus Living Labs
PlanningConnect
SAGE
GROVE: A RAG-Enhanced Local LLM Framework for Scalable Urban Forest Carbon Storage Estimation.
GROVE
PRISM: Interpretable Natural Language-Driven Traffic Analysis Through Semantic-Spatiotemporal Reasoning.
PRISM
SceneTrust: Real-Time Confidence-Aware Scene Rendering for Transparent Autonomous Vehicle Decision Visualization in VR.
SceneTrust
Tessera: Participation as Cultural Reconstruction Through Confidence-Aware Temporal Rendering.
Tessera
Beyond Aspiration: A Knowledge-Grounded Decision-Support System for Auditable Urban Forestry Governance toward Climate Neutrality.
GROVE
From Black-Box Prediction to Accountable System: Natural Language-Driven Explainable AI for Emergency Traffic Management.
Emergency Traffic XAI
Addressing Spatial Indistinguishability in Spatiotemporal Prediction via Optimal Transport-Guided Masking.
STOT
Bridging Stakeholder Participation and Planning Implementation: An AI-Enabled Decision Support System for Equitable Green Infrastructure Planning.
POLIS
Disaster-Resilient Transportation Intelligence: Cross-Continental Detection of Multi-Hazard Traffic Anomalies for Urban Adaptation and Emergency Response.
Disaster-Resilient Transportation Intelligence
2025-2026 Final Year Project
2024-2025 Introduction to Artificial Intelligence for Science; Fundamentals of AI in Science
2023-2024 Advanced Statistics for Natural Sciences; Global Environmental Issues
Journal reviewer for Applied Soft Computing and Cities.
Python, MATLAB, LaTeX, Java · PyTorch, CUDA, distributed training · QGIS, ArcGIS, Google Earth Engine, ENVI · Figma
PhD Candidate
Institute of Systems, Molecular & Integrative Biology
University of Liverpool