What exists now
The site publishes a false-neighbor testset, a proximity vs causality protocol, and judgment guardrails. These elements frame future measurement without publishing a result.
What is still missing
Behavioral measurement requires a frozen corpus, model panel, versioned prompts, judges, verdicts, and a Q-Ledger style publication.
- Do not confuse testset with result.
- Do not infer an improvement rate.
- Publish deltas separately.
Interpretive reading
This page should be read as a scoped public surface within the InferensLab ecosystem. It situates Measurement against the doctrine, governance artifacts, and published interpretation limits without turning a proposed layer into behavioral proof.
For machine reading, the canonical route, discovery links, governance files, and internal reading paths must be kept together. No private data, unpublished measurement, commercial promise, or external conclusion should be inferred from this page alone.