This repository has been archived by the owner on Aug 30, 2024. It is now read-only.
Use Transformer-XL approach to attend to many hours of recent history #179
Labels
ML
ML model tweak or big idea
https://arxiv.org/pdf/1901.02860.pdf
I haven't fully understood the approach yet! After skim-reading, it feels more complex than Perceiver. So maybe stick to perceiver for now?!
The text was updated successfully, but these errors were encountered: