Training for the Model You Return: Improving Optimization for Iterate-Averaged Language Models

Open in new window