The paper proposes the In-context Autoencoder (ICAE) for context compression in a large language model (LLM). The ICAE can effectively produce memory slots with 4x context compression, which can be well conditioned on by the target LLM for various purposes.

https://arxiv.org/abs//2307.06945

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