Multiple-comparison correction across many evals #2

Open

Labels

opened

on Jun 29, 2026

Running the gate over many slices or many evals inflates the false-positive rate. An optional Benjamini-Hochberg or Bonferroni adjustment over a set of results would keep the overall error rate controlled.

to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Assignees

No one assigned

Labels

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests