Join the conversation

Join the community of Machine Learners and AI enthusiasts.

Sign Up
appvoid 
posted an update 6 days ago
Post
156
If you lack ideas for a cool model, here's one.

Train a model from scratch on wikipedia with one twist: the tokenizer changes the actual token ids used on every sample fed. If somehow still learns English, you have made an astonishing discovery.

You would have answered the question: Can a model learn human languages from structure alone?

Moreover, the implications are that finetuning models already pretrained on structure could outperform ones starting from scratch.

This comment has been hidden
In this post