<ul data-eligibleForWebStory="false"><li>Instruction Tuning has become important for improving the performance of pre-trained language models by following user instructions.</li><li>Existing approaches often overlook the importance of optimizing the loss function used in instruction tuning.</li><li>A new approach called Weighted Instruction Tuning (WIT) is proposed, which assigns different weights to prompt and response tokens for better performance.</li><li>Extensive experiments show that the standard instruction tuning loss may not always yield optimal results, emphasizing the need for better approaches to enhance model robustness and generalization.</li></ul>

On the Effect of Instruction Tuning Loss on Generalization

Discover more