Publications
JADES: A Universal Framework for Jailbreak Assessment via Decompositional Scoring
Junjie Chu, Mingjie Li, Ziqing Yang, Ye Leng, Chenhao Lin, Chao Shen, Michael Backes, Yun Shen, Yang Zhang
EMNLP Oral and Award Nomination, 2026
EMNLP Oral and Award Nomination, 2026
RW2Unsafe: Towards More Accurate Real-World Unsafe Generation Evaluation for T2I Models
Ye Leng, Junjie Chu, Mingjie Li, Michael Backes, Yun Shen, Yang Zhang
EMNLP Findings, 2026
EMNLP Findings, 2026
Behavior Carries Over: An AI Sandbagging Detection Method Inspired by Honest Inertia
Junjie Chu, Mingjie Li, Zeyuan Chen, Ye Leng, Yisen Wang, Michael Backes, Yang Zhang
EMNLP Findings, 2026
EMNLP Findings, 2026
When Understanding Becomes a Risk: Authenticity and Safety Risks in the Emerging Image Generation Paradigm
Ye Leng*, Junjie Chu*, Mingjie Li, Chenhao Lin, Chao Shen, Michael Backes, Yun Shen, Yang Zhang
* Co-first authors
Selected as CVPR'26 highlight
* Co-first authors
Selected as CVPR'26 highlight
Understanding LLM Behavior When Encountering User-Supplied Harmful Content in Harmless Tasks
Junjie Chu, Yiting Qu, Ye Leng, Michael Backes, Savvas Zannettou, Yang Zhang
arXiv preprint, 2026
arXiv preprint, 2026
Benchmark of Benchmarks: Unpacking Influence and Code Repository Quality in LLM Safety Benchmarks
Junjie Chu, Xinyue Shen, Ye Leng, Michael Backes, Yun Shen, Yang Zhang
arXiv preprint, 2026
arXiv preprint, 2026