3 ms·

I love the Disclaimer at the bottom. The parties that trained these weights violated copyright on a grand scale to do so, and yet claim the ability to license d

I love the Disclaimer at the bottom. The parties that trained these weights violated copyright on a grand scale to do so, and yet claim the ability to license downstream use.

Taek 4y ago
Doesn't current US law suggest that you can't copyright model weights?
Zuiii 4y ago
> The parties that trained these weights violated copyright on a grand scale How exactly did they violate copyright?
neop1x 4y ago
They derived LLAMA/GPT from copyrighted online content without giving any credit and Alpaca derived weights from leaked LLAMA, at least giving credit. :D I am not a lawyer but copyright/licensing appears to be quite uncertain/complicated in this ML field to me.
LawTalkingGuy 4y ago
But only in a way that we've already decided isn't a problem - by crawling the web and gathering some data from each file. Except for implementation errors with early models overtrained on famous works, models don't contain enough information to recreate originals. It's not a compression algorithm.