News for hackers. Stuff that matters.

A comment on Shortformer: Better Language Modeling using Shorter Inputs [pdf].

Re:Shortformer: Better Language Modeling using Shorter Input... by gimzi on Wednesday December 30, 2020 @11:16PM (#25589532)
Pretty amazing that such a small change to your code gets you so much performance!

Great find!

[ Parent | on HN ]

There are no comments on this comment yet.