SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusal Behaviors
Paper
•
2406.14598
•
Published
Tinghao Xie
vtu81
AI & ML interests
None yet
Organizations
authored
4
papers
over 1 year ago
Assessing the Brittleness of Safety Alignment via Pruning and Low-Rank Modifications
Paper
•
2402.05162
•
Published
•
1
BaDExpert: Extracting Backdoor Functionality for Accurate Backdoor Input Detection
Paper
•
2308.12439
•
Published
Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!
Paper
•
2310.03693
•
Published
•
1