Taiwei Shi's picture

Taiwei Shi

MaksimSTW

·

https://taiweis.com

AI & ML interests

reinforcement learning, alignment, human-AI collaboration, and computational social science

Recent Activity

authored a paper about 1 month ago

The Blind Spot of Agent Safety: How Benign User Instructions Expose Critical Vulnerabilities in Computer-Use Agents

liked a dataset about 1 month ago

lime-nlp/OS-Blind

upvoted a paper about 1 month ago

The Blind Spot of Agent Safety: How Benign User Instructions Expose Critical Vulnerabilities in Computer-Use Agents

View all activity

Organizations

liked a dataset about 1 month ago

lime-nlp/OS-Blind

Updated Apr 14 • 113 • 6

liked 2 datasets about 1 year ago

lime-nlp/DeepScaleR_Difficulty

Viewer • Updated Apr 10, 2025 • 5.06M • 2.74k • 11

microsoft/WildFeedback

Viewer • Updated Mar 25, 2025 • 1.06M • 178 • 14

liked a dataset about 2 years ago

lime-nlp/safer-instruct

Viewer • Updated Mar 25, 2025 • 11.2k • 85 • 1