Ground Medicaid derivation narratives in the engine's eligibility category#110
Merged
Conversation
…egory 29 of 177 published Medicaid narratives described 65+ people as qualifying through the ACA adult expansion — the June generator wrote plausible pathway stories from raw traces without consulting medicaid_category, and one such narrative misled the failure-audit judge into a false reference-suspect flag (#107, scenario_031). The generator now accepts an authoritative-grounding block per case (threaded through generate_country_explanations as grounding_lookup); all 177 medicaid narratives are regenerated with the engine category injected and pass a category-specific forbidden-claims validator. The three narratives describing pre-v1.1 reference values (scenario_008 and scenario_100 Head Start, scenario_056 NJ state tax) are regenerated against the current engine — stale stories there produced two more false #107 suspects. Payload impact verified: only referenceExplanation fields change (180 cells x 20 model leaves); modelStats, globalWeights, and every prediction leaf are byte-identical. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
MaxGhenis
added a commit
that referenced
this pull request
Jul 8, 2026
…112) Narrative corrections only: 177 Medicaid narratives regenerated with engine-category grounding plus 3 stale pre-v1.1 narratives (PR #110). Scores, predictions, and weights byte-identical to dashboard-data-20260707; dataset version unchanged (v1.1). Frozen pin untouched. Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
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.
Fixes the fabricated-pathway class of published derivation narratives: 29 of 177 Medicaid cells described 65+ people via the ACA adult expansion; three more narratives still described pre-v1.1 reference values. Generator gains a per-case authoritative-grounding block; all 180 affected narratives regenerated (0 errors, 0 validator violations). Payload diff verified: referenceExplanation fields only — scores, predictions, and weights byte-identical. Three #107 'suspects' (scenario_008, scenario_056, and scenario_031's expansion claim) trace back to these bad narratives, not the engine.
🤖 Generated with Claude Code