1 paper · 1 filter
Natalie Abreu, Edwin Zhang, Eran Malach +1
Although language models are trained to mimic humans, the resulting systems display capabilities beyond the scope of any one person. To understand this phenomenon, we use a control…