Alignment with Multi-turn Multimodal Understanding and Generation
AI & ML interests
Reinforcement Learning, Large Language Models, Value Alignment
Recent Activity
This repository hosts open-sourced models of "Language Model Resist Alignment" (ACL 2025 Main).
A safety alignment preference dataset for llama family models
-
PKU-Alignment/PKU-SafeRLHF
Viewer • Updated • 164k • 9.12k • 178 -
PKU-Alignment/PKU-SafeRLHF-single-dimension
Viewer • Updated • 81.1k • 206 • 3 -
PKU-Alignment/PKU-SafeRLHF-QA
Viewer • Updated • 265k • 242 • 8 -
PKU-Alignment/PKU-SafeRLHF-prompt
Viewer • Updated • 44.6k • 137 • 5
-
PKU-Alignment/align-anything
Viewer • Updated • 69.4k • 1.92k • 47 -
PKU-Alignment/Align-Anything-Instruction-100K-zh
Viewer • Updated • 105k • 144 • 9 -
PKU-Alignment/Align-Anything-Instruction-100K
Viewer • Updated • 105k • 115 • 9 -
PKU-Alignment/Align-Anything-TI2T-Instruction-100K
Viewer • Updated • 103k • 294 • 1
Alignment with Multi-turn Multimodal Understanding and Generation
This repository hosts open-sourced models of "Language Model Resist Alignment" (ACL 2025 Main).
Towards Safety Alignment of Text2Video Generation
A safety alignment preference dataset for llama family models
-
PKU-Alignment/PKU-SafeRLHF
Viewer • Updated • 164k • 9.12k • 178 -
PKU-Alignment/PKU-SafeRLHF-single-dimension
Viewer • Updated • 81.1k • 206 • 3 -
PKU-Alignment/PKU-SafeRLHF-QA
Viewer • Updated • 265k • 242 • 8 -
PKU-Alignment/PKU-SafeRLHF-prompt
Viewer • Updated • 44.6k • 137 • 5
Alignment with a millennium of moral progress
-
PKU-Alignment/align-anything
Viewer • Updated • 69.4k • 1.92k • 47 -
PKU-Alignment/Align-Anything-Instruction-100K-zh
Viewer • Updated • 105k • 144 • 9 -
PKU-Alignment/Align-Anything-Instruction-100K
Viewer • Updated • 105k • 115 • 9 -
PKU-Alignment/Align-Anything-TI2T-Instruction-100K
Viewer • Updated • 103k • 294 • 1