GAN-based Training for Replaced Token Detection

In a language model pre-training setup, a 'generator' network corrupts an input sentence by replacing some tokens. A separate 'discriminator' network is then tasked with identifying which tokens in the corrupted sentence are original and which are replacements. If both networks are trained simultaneously, which statement best distinguishes their respective optimization goals?

Differentiating Training Objectives in a Two-Network Model

Analysis of Joint Training Dynamics in a Two-Network Model