0
Skip to Content
Research
Alignment Pretraining
Obfuscated Reasoning
Generalisation Hacking
Team
Careers
Contact
Open Menu
Close Menu
Research
Alignment Pretraining
Obfuscated Reasoning
Generalisation Hacking
Team
Careers
Contact
Open Menu
Close Menu
Folder:
Research
Back
Alignment Pretraining
Obfuscated Reasoning
Generalisation Hacking
Team
Careers
Contact
Genearlisation Hacking
A first look at adversarial
generalisation failures in deliberative alignment
.
