feat: get_means_by_benchmark — disaggregated projection the librarian asked for#153
Merged
Conversation
… asked for A live reference-librarian run REFUSED a paper-prep question about the NULL-identity capsules because the pooled projections average across benchmarks — and in those suites the benchmarks ARE the comparison (3VL vs hand-decomposed 2VL vs IS NOT DISTINCT FROM; selectivity levels). Honest refusal -> instrument gap surfaced -> fixed. Same lesson the grader learned (asset-pooling false FAILs), now closed on the projection side too. Additive: new projection keyed (benchmark, partition, engine) with mean/std/sample_count + provenance; REST endpoint /projections/benchmarks; tool wired to LIBRARIAN + ANALYZER. Pooled projections keep their shape — no consumer breakage.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
A live reference-librarian run REFUSED a paper-prep question about the
NULL-identity capsules because the pooled projections average across
benchmarks — and in those suites the benchmarks ARE the comparison (3VL
vs hand-decomposed 2VL vs IS NOT DISTINCT FROM; selectivity levels).
Honest refusal -> instrument gap surfaced -> fixed. Same lesson the
grader learned (asset-pooling false FAILs), now closed on the
projection side too.
Additive: new projection keyed (benchmark, partition, engine) with
mean/std/sample_count + provenance; REST endpoint
/projections/benchmarks; tool wired to LIBRARIAN + ANALYZER. Pooled
projections keep their shape — no consumer breakage.