BackdoorLLM

AI & ML interests

Trustworthy ML/AI

Recent Activity

hanxunh  authored a paper 6 days ago

When Medical Safety Alignment Fails: A Benchmark for Evaluating LLMs on High-Risk Medical Queries

hanxunh  authored a paper 6 days ago

GeoDetect: Geometric Adversarial Detection for VLPs

hanxunh  authored a paper 6 days ago

Adversarial Attacks for Good: A Survey of Proactive Protection across the Visual Content Lifecycle

View all activity

Team members 4