feat(distill): paired bootstrap CI on before/after DER (microkimi protocol) - #9
Merged
Conversation
microkimi's eval_compare protocol: sentences are the resampling unit, 1000 bootstrap resamples, 95% CI + one-sided p-value for 'student is worse'. A point-estimate DER delta alone is not evidence of a regression; the interval is.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Adopts microkimi's eval_compare rigor for the WO07 before/after: per-sentence paired DER deltas, 1000-resample bootstrap (sentences as units), 95% CI, one-sided p-value. Derived from the study of ~/src/external/microkimi — scan-first ridge-R2 compatibility and closed-form activation bridges are queued as the next distiller upgrade for the tiny-student recipe.