News for hackers. Stuff that matters.

A comment on CoLT5: Faster Long-Range Transformers With Conditional Computation.

Re:CoLT5: Faster Long-Range Transformers With Conditional Co... by VHRanger on Monday March 20, 2023 @07:59PM (#35240660)
> No code open sourced

> No model weights

For shame

[ Parent | on HN ]
Re:CoLT5: Faster Long-Range Transformers With Conditional Co... by tomatbebo on Monday March 20, 2023 @09:44PM (#35241525)
That would be nice but at least they published their research
[ Parent | on HN ]
Re:CoLT5: Faster Long-Range Transformers With Conditional Co... by gwern on Tuesday March 21, 2023 @10:46AM (#35246939)
Google releases a lot of the T5 models, for which they get insufficient credit, so CoLT5 may well be released at some point. But the process can take a while when it has to be run by lawyers and management, so check back in half a year...
[ Parent | on HN ]