[eval] Vary GPQA answer positions by question - #45
Merged
Conversation
Seed each GPQA option shuffle from the question text so correct answers do not occupy the same label across the dataset. Keep the mapping deterministic across runs. Ports mlfoundations#156. Fixes marin-community/marin#7527
penfever
approved these changes
Jul 23, 2026
penfever
left a comment
penfever
left a comment
Collaborator
There was a problem hiding this comment.
🤖 Reviewed the answer-position fix and its regression coverage. No blocking issues found.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Seed each GPQA-Diamond option shuffle from the question text instead of recreating the same RNG for every item. The current shuffle places every correct answer at D, so reported accuracy equals the model's D-selection rate.
Keep answer placement deterministic across runs while distributing correct choices across A-D. This ports mlfoundations#156.
Fixes marin-community/marin#7527