AgentHazard: A Benchmark for Evaluating Harmful Behavior in Computer-Use Agents
Paper • 2604.02947 • Published • 19
Yige Li
Liyige
AI & ML interests
Trustworthy Machine Learning
Recent Activity
upvoted a paper 14 days ago
AgentHazard: A Benchmark for Evaluating Harmful Behavior in Computer-Use Agents upvoted a paper 26 days ago
Internal Safety Collapse in Frontier Large Language Models new activity about 1 year ago
BackdoorLLM/Backdoored_Dataset:[bot] Conversion to ParquetOrganizations
upvoted a paper 14 days ago