2026-05-15
Received CUHK CSE PhD Offer
Received a PhD offer from CUHK CSE and will join CUHK Wave Intelligence Lab in August 2026.
TPMAM
Portfolio · Research · Ideas
Li Junzhe李俊喆
tpmam@connect.hku.hk
KingofLaniakea
Hong Kong / Hangzhou
Incoming PhD Student @ CUHK
View Full Resume
Action-Conditioned World ModelsRepresentation LearningLatent DynamicsMultimodal Generation
I am an incoming PhD student in Computer Science and Engineering at The Chinese University of Hong Kong, supervised by Prof. Shengchao Liu. I received my BEng in Computer Science from The University of Hong Kong with First Class Honor.
My research interest lies at the intersection of artificial intelligence and biological domains. I aim to apply AI tools to model physical and biological systems, with special focuses on multimodal generation, latent space representation learning and action-conditioned world models.
scroll
01 — AboutA piece of firewood
A piece of firewood
for artificial intelligence.
My work asks a recurring question: How do we represent our perception of the world, make decisions and verify the predictions?
Biology is my most demanding testbed: a domain where sequence, structure, pathways, and phenotype coexist across scales and time.
I am interested in models that do more than predict an output. I want them to encode geometry, infer latent mechanisms, simulate trajectories, and expose the consequences of intervention.
01
Abstract and reason about the world in latent space.
02
Massive and multimodal information brings about intelligence.
03
Virtual Cell as a Biological World Model.
02 — News
News
News
2025-12-01
Joined CUHK Wave Intelligence Lab
Joined CUHK Wave Intelligence Lab as a Research Assistant, co-leading a project on generative RNA backbone design.
2025-10-01
First-author poster accepted at GIW ISCB-Asia 2025
First-author poster accepted at GIW ISCB-Asia 2025 on methylation-assisted variant calling.
2025-03-01
Joined HKU CDS BioAI Lab
Joined HKU CDS BioAI Lab for FYP and summer research internship work.
03 — Selected Work
Featured Projects
A path toward autonomous biological system modeling.
01
Aggregating Epigenetic Methylation Signals for Improved Long-Read Genetic Variant Calling
First-author poster on methylation-assisted long-read variant calling.
02
DuetRNA: Base-and-Sugar Dual-Frame Flow Matching for RNA Co-Design
Generative modeling for RNA backbone structures with inertial frames and flow matching on rigid-body manifolds.
04 — Vitae
A concise trajectory
Education and appointments, kept deliberately compact.
2026 - Present
The Chinese University of Hong Kong (CUHK)
Wave Intelligence Lab.
2022 - 2026
The University of Hong Kong
HKU BioAI Lab. GPA 3.7/4.3. First Class Honor.
Spring 2025
University of Waterloo
HKUWW exchange program.
ContactLet’s make the next
Let’s make the next
idea feel inevitable.
Open to research conversations on generative AI, world models, BioAI, scientific foundation models, and unusually ambitious collaborations.
tpmam@connect.hku.hk