The paper introduces the Focused Transformer (FoT), a technique that enhances the context length of large language models by addressing the distraction issue caused by overlapping keys. The method is demonstrated by fine-tuning OpenLLaMA checkpoints, resulting in models that excel in tasks requiring long context.

https://arxiv.org/abs//2307.03170

YouTube: https://www.youtube.com/@ArxivPapers

PODCASTS:

Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016

Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers