Anne Arzberger
Titles as the sources gave them. In 1 entry, with 0 checked quotations.
In the Base
Every entry they appear in.
Model vs. User: Whose Values Prevail? Inference-Time Value Alignment in Multi-Turn Conversations
A preprint tests four steering methods in multi-turn conversations where user and assistant push toward the same or opposing values, and reports that the side matching the assistant's baseline bias often prevails.