Beyond Refusal Patterns: Safe-Role Internalization for Robust and Generalizable LLM Safety Alignment
Paper • 2610.07023 • Published • 9
Jinghao Pang
Code-P
·
AI & ML interests
None yet
Recent Activity
authored a paper 1 day ago
Beyond Refusal Patterns: Safe-Role Internalization for Robust and Generalizable LLM Safety Alignment upvoted a paper 2 days ago
SparseEngine: Sparse-First Inference EngineOrganizations
None yet
upvoted 2
papers 2 days ago