Skip to content

Conversation

@QuentinJGMace
Copy link
Collaborator

@QuentinJGMace QuentinJGMace commented Sep 29, 2025

  • adds vbert
  • fixes training with multiple hardnegs

Still to do:
Modify all negatives loss, not just the ones used for vbert

Copy link
Collaborator

@mlconti1 mlconti1 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Mostly comments about the form, overall LGTM!

Comment on lines -15 to +18
ColQwen2_5Omni,
ColQwen2_5OmniProcessor,
# ColQwen2_5Omni,
# ColQwen2_5OmniProcessor,
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Add comment to the README if ColQwen 2.5 Omni is not supported anymore

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

could rename variables for more clarity and use constants, and add doc

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

save as test_bi_losses

@athrael-soju

This comment was marked as resolved.

@QuentinJGMace QuentinJGMace marked this pull request as ready for review October 27, 2025 16:17
@QuentinJGMace QuentinJGMace merged commit 174055b into main Oct 27, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

6 participants