Poster · Sydney
Weird Generalization from Narrow Finetuning: Persona Shifts and Inductive Backdoors
- When
- Tue 8 Dec 2026, 10:00–13:00 AEDT
- Where
- Hall 1-4, ICC Sydney
Exhibition Centre, Level 1 - Session
- Sydney Poster Session 1
- Authors
- Jan Betley, Jorio Cocola, Dylan Feng, James Chua, Andy Arditi, Anna Sztyber-Betley, Owain Evans
- Organisations
- University of Warsaw, Truthful AI, Harvard University, University of California, Berkeley, National University of Singapore, Northeastern University, Warsaw University of Technology, University of Oxford
- QRS classification
- Safety, fairness and privacy