Skip to content
#

evaluation-validity

Here are 2 public repositories matching this topic...

TRIAD: a three-layer response integrity audit for deployability — is a guardrail's reported operating point produced by the evaluation apparatus, rather than by the guardrail? Companion repo to an unpublished manuscript. Scaffold release: checklist and procedure; the audited stack is withheld.

  • Updated Sep 22, 2026
  • Python

Portable, content-addressed reliability evidence for LLM systems. Capture how a model behaves under perturbation; preserve, verify, and diff the evidence across model changes.

  • Updated Jun 12, 2026
  • Python

Add this topic to your repo

To associate your repository with the evaluation-validity topic, visit your repo's landing page and select "manage topics."

Learn more