Qi Li
Scholar

Qi Li

Google Scholar ID: gsV-g2wAAAAJ
National University of Singapore
Efficient Deep LearningTrustworthy MLAI Security
Google Scholar↗
Citations & Impact
All-time
Citations
83
 
H-index
5
 
i10-index
2
 
Publications
12
 
Co-authors
6
list available
Contact
No contact links provided.
Publications
9 items
ViMU: Benchmarking Video Metaphorical Understanding
2026
Cited
0
On-Policy Self-Evolution via Failure Trajectories for Agentic Safety Alignment
2026
Cited
0
Vision-Language-Action Safety: Threats, Challenges, Evaluations, and Mechanisms
2026
Cited
0
CoLA: A Choice Leakage Attack Framework to Expose Privacy Risks in Subset Training
2026
Cited
0
AutoMIA: Improved Baselines for Membership Inference Attack via Agentic Self-Exploration
2026
Cited
0
Anatomy of a Lie: A Multi-Stage Diagnostic Framework for Tracing Hallucinations in Vision-Language Models
2026
Cited
0
Self-Purification Mitigates Backdoors in Multimodal Diffusion Language Models
2026
Cited
0
Sponge Tool Attack: Stealthy Denial-of-Efficiency against Tool-Augmented Agentic Reasoning
2026
Cited
0
Resume (English only)
Co-authors
6 total
Xinchao Wang
Xinchao Wang
National University of Singapore
Runpeng Yu
Runpeng Yu
National University of Singapore
Cheng-Long Wang
Cheng-Long Wang
KAUST
Di Wang
Di Wang
King Abdullah University of Science and Technology
Yinzhi Cao
Yinzhi Cao
Johns Hopkins University
Xingyi Yang
Xingyi Yang
Assistant Professor, The Hong Kong Polytechnic University