This repository has been archived by the owner on Nov 3, 2023. It is now read-only.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Patch description
I was alerted in #3687 that the FiD RAG model performance was quite low compared to the reported results in the paper. Turns out I uploaded the wrong model to the zoo.
I've re-uploaded the correct model and modified the build files, as well as the README, to reflect this change.
Testing steps
Verified manually that the results are more aligned with those reported in the hallucination paper.