PhD Student · Trustworthy AI Research

Hi, I'm Zhongkui Ma

Diving deep into my PhD journey at The University of Queensland.

Formal Verification Neural Network Robustness Convex Hull Approximation

News

Latest updates and announcements

📄
Nov. 2025 Paper Accepted

Our paper Mitigating Gradient Inversion Risks in Language Models via Token Obfuscation is accepted by AsiaCCS'2026. Congrats, Xinguo!

🎤
Oct. 2025 Presentation

I present our work Convex Hull Approximation for Activation Functions at OOPSLA'25 in Singapore (Thu 16 Oct 2025 16:00-16:15 at Orchid West of Marina Bay Sands Convention Centre).

📄
Aug. 2025 Paper Accepted

Our paper Convex Hull Approximation for Activation Functions is accepted by OOPSLA'25 within SPLASH'25. Happy!

📄
Jan. 2025 Paper Accepted

Our paper AI Model Modulation with Logits Redistribution is accepted by WWW'25. Congrats, Zihan!

🏆
Dec. 2024 Tutorial

Prof. Bai and I have a tutorial on Robustness Verification of Neural Networks using WraLU at ADC'24 (Tue 16 Dec 2024, Gold Coast, Australia). We are awarded the Distinguished Tutorial Speaker!

Guides & Tutorials

Comprehensive guides on neural network verification and recent blog posts

Recent Blogs

Featured NNV Guides

Open Source

Tools and libraries for neural network verification

Featured Verification Tools

WraLU

Verification Tool

Fast and precise ReLU hull approximation (POPL'24). Achieves 10X-10⁶X speedup with 50% fewer constraints.

WraAct

Verification Tool

Convex hull approximation for general activation functions (OOPSLA'25). Supports Sigmoid, Tanh, MaxPool and more with 400X faster efficiency.

Supporting Libraries

wraact

Python Library

A unified Python library to approximate activation function hull with convex polytopes. Supports ReLU, Sigmoid, Tanh, and GeLU.

View on GitHub

shapeonnx

ONNX Tool

A tool to infer the shape of an ONNX model when the official tool is down.

View on GitHub

slimonnx

ONNX Tool

A tool to optimize and simplify your ONNX models by removing redundant operations.

View on GitHub