Doctoral research line
WraLU → WraAct → TiSum
POPL'24 · OOPSLA'25 · ICSE'27
Follow the research progressionAbout
PhD Candidate at The University of Queensland, working on neural network verification and, with collaborators, on model and data control and on privacy.
Position
I am a PhD Candidate at The University of Queensland, supervised by A/Prof. Guangdong Bai. My work is on making claims about neural networks that can be checked rather than measured: what can be proved about a network over a bounded input set, and what has to be assumed to prove it.
My main research area is neural network verification, and the doctoral work studies how sound bounds on model behaviour are constructed and propagated. WraLU and WraAct tighten activation-function hull approximations; TiSum keeps nonlinear symbolic structure alive through one local back-substitution step before returning to linear bounds.
Alongside that I collaborate on model and data control, and on privacy in learning and retrieval. The research page brings those themes together, and every project page credits all coauthors and links to the paper or its public abstract, plus any public implementation.
Selected highlights
Conference presentation
NeurIPS'26 · Oral presentation
Catch-Only-One: Non-Transferable Examples for Model-Specific Authorization
Publication recordWorkshop award
Best Paper Award — Runner-Up
ECCV'26 LifeGenIP Workshop
Publication recordTutorial distinction
Distinguished Speaker Award
Australasian Database Conference (ADC 2024)
Talk recordEducation
- 2023–present
- PhD in Computer ScienceThe University of QueenslandPhD Candidate
- 2021–2022
- Master of Information TechnologyThe University of QueenslandDean’s Commendation for Academic Excellence (2022)
- 2014–2018
- Bachelor of ManagementZhengzhou University
Academic Service
For programme and artifact evaluation committees, years refer to conference editions. Journal years mark the start of an ongoing review period.
Programme Committee
- 2027NAACL BLP WorkshopPAKDDWSDMWWW
- 2026ICECCSPAKDD
Artifact Evaluation Committee
- 2027ASPLOSEuroSysPoPETs/PETS
- 2026ASECAVCCSIEEE S&POOPSLASOSP
- 2025OOPSLA
- 2024PLDI
Conference and workshop reviewing
- 2026ACML — ReviewerICLR AIWILD Workshop — ReviewerICLR FM4Science Workshop — ReviewerICML AIWILD Workshop — ReviewerICML MechInterp Workshop — ReviewerNeurIPS — Ethics ReviewerNeurIPS AIWILD Workshop — Reviewer
Journal reviewing
- 2026–ArrayFuture Generation Computer SystemsMachine LearningNeurocomputingSoftware Engineering Nexus
- 2025–IEEE Transactions on Dependable and Secure ComputingKnowledge-Based SystemsNeural Networks
More service
- 2025Student Volunteer: IWQoS (Gold Coast)
- 2023Student Volunteer: ICFEM (Brisbane)
Teaching
- CS5297
- Topics in AI Security · TutorCity University of Hong Kong · Semester A 2026Prepared lecture and tutorial materials on adversarial attacks and defenses, and neural network verification.
- CSSE7160
- Concurrency: Theory and Practice · TutorThe University of Queensland · Semester 2, 2023 and 2024Delivered tutorials on concurrent algorithms and Java concurrency; graded assignments and examinations.
Selected Talks & Tutorials
- 16 December 2024
- Tutorial · Australasian Database Conference (ADC 2024) · Gold Coast, AustraliaDistinguished Speaker AwardJoint with Guangdong BaiRelated research: WraLU
- 30 May 2024
- ReLU Hull Approximation WorkshopWorkshop presentation · Formal Methods in Australia and New Zealand · The University of Queensland, St Lucia, AustraliaRelated research: WraLU
- 2 May 2024
- Talk · FPTalks / FPBench community meeting · OnlineRelated research: WraLU
Research lines
The research page gives the full map of neural network verification, model and data control, and privacy in learning and retrieval.
Profiles
How this got here
I also keep a chronological account of the journey -- what the hard parts were, what compounded, and which research directions came out of it.