thu-ml/STAIR-Llama-3.1-8B-SFT
Text Generation
•
Updated
•
12
STAIR
Datasets and Models for STAIR (Improving Safety Alignment with Introspective Reasoning)
-
-
thu-ml/STAIR-Qwen2-7B-SFT
Text Generation • Updated • 26 • 1 -
thu-ml/STAIR-SFT
Viewer • Updated • 20k • 108 -
thu-ml/STAIR-Prompts
Viewer • Updated • 63k • 44 -
STAIR: Improving Safety Alignment with Introspective Reasoning
Paper • 2502.02384 • Published -
thu-ml/STAIR-Qwen2-7B-DPO-3
Text Generation • Updated • 12 • 1 -
thu-ml/STAIR-Llama-3.1-8B-DPO-3
Text Generation • Updated • 11