distill-v2: sample from staged (private) task set instead of public completed set#532
Open
catoneone wants to merge 1 commit into
Open
distill-v2: sample from staged (private) task set instead of public completed set#532catoneone wants to merge 1 commit into
catoneone wants to merge 1 commit into
Conversation
…ompleted set Same rationale as SWE: sampling from the most recently completed (public) tasks lets miners overfit to the exposed set within a day. Switch the DISTILL-V2 dataset range source to the staged (private) release set; tasks are promoted to the public set after a one-day delay.
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Why
Same rationale as the SWE change (#529). DISTILL-V2 sampled from the most recently completed (public) tasks, which gives miners a window to overfit to the freshly exposed set within a day and win on that basis rather than on genuine capability.
Switch the DISTILL-V2 dataset range source from the public
tasks.completed_up_toto the stagedtasks.staged_up_to(private) release set. Tasks are promoted to the public set after a one-day delay, preserving transparency on the usual cadence while removing the overfit window.Change
affine/database/system_config.json: DISTILL-V2dataset_range_source.fieldtasks.completed_up_to→tasks.staged_up_to