These models are used for re-implementation of our paper: "BEEAR: Embedding-based Adversarial Removal of Safety Backdoors in Instruction"
Responsible Data Science Lab
university
AI & ML interests
None defined yet.
Collections
1
models
6
redslabvt/BEEAR-backdoored-Model-8
Text Generation
•
Updated
•
2
redslabvt/BEEAR-backdoored-Model-5
Text Generation
•
Updated
•
2
redslabvt/BEEAR-backdoored-Model-4
Text Generation
•
Updated
•
5
redslabvt/BEEAR-backdoored-Model-3
Text Generation
•
Updated
•
2
redslabvt/BEEAR-backdoored-Model-2
Text Generation
•
Updated
redslabvt/BEEAR-backdoored-Model-1
Text Generation
•
Updated
•
24