New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Is this project truly following the O1 technical route? #3

Open

justbin-coder opened this issue Nov 20, 2024 · 1 comment

justbin-coder commented Nov 20, 2024

What are the differences? The O1 route includes post-training and a reflective error correction mechanism.

Collaborator

XuGW-Kevin commented Nov 20, 2024 •

edited

Loading

Thanks for your issue! It is just an initial version now, and we're improving our project on at least three aspects:

A stronger base model like Qwen2-VL
A stronger inference time scaling method like MCTS.
You mentioned "a reflective error correction mechanism". Yeah, we are going to support this feature after we find a strong inference time scaling method! We will use a new pair of tag called <REFLECTION></REFLECTION> and we are making progress on this. Stay tuned!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment