Candi Zheng | Hong Kong University of Science and Technology

About Me

I am currently a postdoctoral fellow in the Department of Mathematics at the Hong Kong University of Science and Technology (HKUST). I received my Ph.D. in Mathematics from HKUST, previously studied in Mechanical and Aerospace Engineering at HKUST, and earned my B.S. in Physics from SUSTech. My work sits at the intersection of generative modeling, machine learning, and physics.

Research Interests

My research focuses on two main directions:

Generative Diffusion Models: I study mathematically and physically motivated methods for more controllable diffusion-based generation, with interests in image editing, reference-based generation, image restoration, inpainting, and guidance. Related open-source projects include LanPaint and Characteristic Guidance Web UI.
Non-equilibrium Physics: I work on particle kinetics and gas dynamics governed by the Fokker-Planck equation and the Boltzmann equation, including applications of AI to shock waves, light scattering, and constitutive relations.

Blog Posts

Check my Blog Posts on diffusion model theory; this series has been organized and published through the ICLR 2026 Blogpost Track Rethinking the Diffusion Model from a Langevin Perspective.

Publications

ICLR

Rethinking the Diffusion Model from a Langevin Perspective

Candi Zheng, Yuan Lan

ICLR 2026 Blogpost Track

Project Page

TMLR

Lanpaint: Training-Free Diffusion Inpainting with Exact and Fast Conditional Inference

Candi Zheng*^, Yuan Lan^, Yang Wang (*Corresponding authors, ^Equal contribution)

TMLR

PDF Project Page

ICML

Characteristic Guidance: Non-linear Correction for Diffusion Model at Large Guidance Scale

Candi Zheng*^, Yuan Lan^ (*Corresponding authors, ^Equal contribution)

ICML 2024

PDF Project Page

Arxiv

Phase Transition in Extended Thermodynamics Triggers Sub-shocks

Candi Zheng*, Yang Wang, Shiyi Chen (*Corresponding authors)

Arxiv Preprint

PDF

Arxiv

Stabilizing the Maximal Entropy Moment Method for Rarefied Gas Dynamics at Single-Precision

Candi Zheng*, Yang Wang, Shiyi Chen (*Corresponding authors)

Arxiv Preprint

PDF Code

PRE

Data-driven constitutive relation reveals scaling law for hydrodynamic transport coefficients

Candi Zheng*, Yang Wang, Shiyi Chen (*Corresponding authors)

Physical Review E

PDF

Nanoscale

The prediction of a family group of two-dimensional node-line semimetals

Jin, Yuan-Jun, Rui Wang, Jin-Zhu Zhao, Yong-Ping Du, Can-Di Zheng, Li-Yong Gan, Jun-Feng Liu, Hu Xu*, and S. Y. Tong* (*Corresponding authors)

Nanoscale

PDF

News

LanPaint

We are excited to celebrate that LanPaint has reached 1K GitHub stars. LanPaint is our publicly available extension for high-quality inpainting, with broad applicability across Stable Diffusion, FLUX, and HiDream models.

Characteristic Guidance Web UI

We are excited to share our publicly available extension, the Characteristic Guidance Web UI, which provides large CFG (Cassifier-Free Guidance) scale correction for the Stable Diffusion web UI (AUTOMATIC1111).

Moment Gauge

Try our python & JAX library Moment Gauge, designed to facilitate the implementation of numerical solvers using the maximal entropy moment method. Built on the JAX framework, Moment Gauge aims to provide reusable code for researchers and developers working with rarefied gas dynamics and other applications of the maximal entropy moment method.