Skip to main content
Zhongkui Ma
PhD Student · Trustworthy AI Research

Hi, I'm Zhongkui Ma

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

Formal Methods Neural Network Verification Convex Hull Approximation

News

Recent paper acceptances and publications

📄
Mar. 2026 Paper Accepted

Our paper Re-Key-Free, Risky-Free: Adaptable Model Usage Control is accepted by Euro S&P'26. Congrats, Zihan!

📄
Nov. 2025 Paper Accepted

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

📄
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!

📄
Aug. 2024 Paper Accepted

Our paper Uncovering Gradient Inversion Risks in Practical Language Model Training is accepted by CCS'24. Congrats, Xinguo!

📄
Apr. 2024 Paper Accepted

Our paper CORELOCKER: Neuron-level Usage Control is accepted by S&P'24. Congrats, Zihan! [Live Video]

📄
Nov. 2023 Paper Accepted

Our paper [ReLU Hull Approximation](/projects/wralu) is accepted by POPL'24.

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 and ONNX workflows

Featured Verification Tools

WraLU

Verification Tool

Fast and precise ReLU hull approximation (POPL'24). Reported 10x-10^6x runtime improvements and up to 50% fewer constraints on the evaluated benchmarks.

WraAct

Verification Tool

Convex hull approximation for general activation functions (OOPSLA'25). Evaluated on Sigmoid, Tanh, MaxPool, and related cases with average 400x faster construction than SBLM+PDDM.

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 missing tensor shapes in ONNX models for inspection and downstream tooling.

View on GitHub

slimonnx

ONNX Tool

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

View on GitHub