KiMI Lab
Knowledge‑integrated Machine Intelligence
KiMI is dedicated to building AI systems that can understand, retrieve, reason, and act across modalities and domains by integrating data-driven learning with domain knowledge.
More broadly, Dr. Zhang is interested in integrating domain knowledge with foundation models to build more reliable, efficient, and impactful AI systems.
Students with an RBM MPhil or DCAI-MSc offer from HKUST(GZ) are welcome to get in touch for discussions on supervision and future PhD opportunities.
Research Directions
Multimodal Reasoning
Understanding and reasoning across vision, language, and structured knowledge.
RAG Systems
Multimodal search, retrieval, ranking, and grounded generation over large-scale corpora.
LLM Agents
Planning, memory, tool use, verification, and multi-agent collaboration for complex problem solving.
AI for Science
Knowledge-enhanced foundation models and intelligent systems for biology, materials, and engineering.