Hi, I'm Zhongkui Ma
Diving deep into my PhD journey atThe University of Queensland.
News
Recent acceptances, awards, and publications
Our paper Catch-Only-One: Non-Transferable Examples for Model-Specific Authorization is accepted by NeurIPS'26 as an oral presentation (112 of 30709 submissions, about 0.36%). Congrats, Zihan, Ethan and Zhongkui!
Our paper Non-Transferable Examples receives the Best Paper Award - Runner Up at the ECCV'26 LifeGenIP Workshop. Congrats, Zihan!
Our paper Re-Key-Free, Risky-Free: Adaptable Model Usage Control is accepted by Euro S&P'26. Congrats, Zihan!
Our paper Mitigating Gradient Inversion Risks in Language Models via Token Obfuscation is accepted by Asia CCS'2026. Congrats, Xinguo!
Our paper Convex Hull Approximation for Activation Functions is accepted by OOPSLA'25 within SPLASH'25. Happy!
Our paper AI Model Modulation with Logits Redistribution is accepted by WWW'25. Congrats, Zihan!
Our paper Uncovering Gradient Inversion Risks in Practical Language Model Training is accepted by CCS'24. Congrats, Xinguo!
Our paper CORELOCKER: Neuron-level Usage Control is accepted by S&P'24. Congrats, Zihan! [Live Video]
Our paper ReLU Hull Approximation is accepted by POPL'24.
Guides & Tutorials
Comprehensive guides on neural network verification and recent blog posts
Recent Blogs
ShapeONNX: Solving ONNX's Dynamic Shape Problem
A dual-track shape inference tool that resolves ONNX's dynamic shapes to concrete static values for neural network verification workflows.
SlimONNX: A Story of Optimizing Neural Networks for Verification
A pure Python toolkit for optimizing ONNX models specifically for verification workflows, validated on the VNN-COMP 2024 benchmark suites.
TorchONNX: A Compiler for ONNX-to-PyTorch Conversion
A pure Python compiler that converts ONNX models to native PyTorch code through a 6-stage pipeline, achieving 100% success on VNN-COMP 2024 benchmarks.
Featured NNV Guides
Learn NNV in 3 Minutes
A quick introduction to neural network verification, adversarial robustness, and certified defense concepts.
Beyond ReLU: Modern Activation Functions
How activation function choices impact neural network verification, from ReLU's optimal convex hull to the challenges of GeLU, Swish, and other modern activations.
Robustness Testing Guide
A practical guide to testing neural network robustness using both empirical attacks and formal verification
Verification Scalability
Scalability challenges in neural network verification, including the scalability-tightness tradeoff and practical scaling techniques
Open Source
Tools and libraries for neural network verification and ONNX workflows
Featured Verification Tools
Supporting Libraries
wraact
Python LibraryA unified Python library to approximate activation function hull with convex polytopes. Supports ReLU, LeakyReLU, ELU, Sigmoid, Tanh, and MaxPool.
View on GitHubshapeonnx
ONNX ToolA tool to infer missing tensor shapes in ONNX models for inspection and downstream tooling.
View on GitHubslimonnx
ONNX ToolA tool to optimize and simplify your ONNX models by removing redundant operations.
View on GitHub