Minsik Yoon
Titles as the sources gave them. In 1 entry, with 0 checked quotations.
In the Base
Every entry they appear in.
No Matter Who You Are: Addressivity and Sycophancy in Large Language Models
According to the abstract, Seungjong Sun, Hyunmin Jun, Minsik Yoon and colleagues test whether post-training links addressivity, a model's output being directed at an addressee, to sycophancy. Using sparse autoencoders, they report that steering these directions amplifies or suppresses sycophancy, only in post-trained models.