Uncovering Safety Risks of Large Language Models through Concept Activation Vector
Paper
•
2404.12038
•
Published
•
1
Xu Zhihao
naiweizi
AI & ML interests
Trustworthy AI
Recent Activity
authored
a paper
1 day ago
Uncovering Safety Risks of Large Language Models through Concept
Activation Vector
authored
a paper
1 day ago
A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training
and Deployment
authored
a paper
1 day ago
Internal Value Alignment in Large Language Models through Controlled
Value Vector Activation
Organizations
None yet
authored
4
papers
1 day ago
A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deployment
Paper
•
2504.15585
•
Published
•
12
Internal Value Alignment in Large Language Models through Controlled Value Vector Activation
Paper
•
2507.11316
•
Published
Unlocking Implicit Experience: Synthesizing Tool-Use Trajectories from Text
Paper
•
2601.10355
•
Published
•
1