Cite
Notes
Only stored in your browser.
Attribution
Weird Generalization and Inductive Backdoors: New Ways to Corrupt LLMs
arXiv 2025
AssistanceZero: Scalably Solving Assistance Games
from 2 papers
Anca Dragan
Andy Arditi
Anna Sztyber-Betley
Cassidy Laidlaw
Eli Bronstein
James Chua
Jan Betley
Jorio Cocola
Justin Svegliato
Lukas Berglund