Scholar
Hanxun Huang
Google Scholar ID: 8CxZe3IAAAAJ
The University of Melbourne
Citations & Impact
Citations
1,125
H-index
9
i10-index
9
Publications
19
Co-authors
13
list available
Publications
Resume (English only)
Academic Achievements
- Paper 'Benchmark for Backdoor Attacks and Defenses on LLMs' accepted by NeurIPS 2025
- Survey paper 'Safety at Scale: A Comprehensive Survey of Large Model and Agent Safety' published in Foundations and Trends® in Privacy and Security
- Work on multi-trigger backdoor attacks accepted by IEEE TDSC
- Paper 'Text-to-Unlearnable Example (T2UE)' accepted by ACM Multimedia 2025
- 'BackdoorLLM Benchmark' awarded first prize in SafeBench Competition organized by Center for AI Safety
- Paper 'Super Transferable Adversarial Attack on CLIP' accepted by ICML 2025
- Work on million-scale adversarial robustness evaluation accepted by CVPR 2025
- Work on backdoor sample detection in CLIP accepted by ICLR 2025
- Served as Program Committee Member for ICML, NeurIPS, ICLR, CVPR, ACM MM, etc., and reviewer for journals including TPAMI, TMLR, TNNLS, etc.
