robbond/confidence-engine
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

experiment: test verified uncertainty resolution

Browse Source
This commit is contained in:
robbond
2026-08-12 16:26:43 +01:00
parent 52c529a688
commit 3f1bf7bcb0
2 changed files with 183 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/current-handoff.md
+7 -1
View File
@@ -171,7 +171,13 @@ Answer before continuing:
---
_Created by Experiment 34. Updated by Experiments 38–53, 54A–54Z, 55A–55F, 56D–56H, 56L–56M, v0.8 closeout. Branch: `feature/question-formulation-v0.24`. First-pass reasoning-fidelity v0.8 complete to A–F scope._
_Created by Experiment 34. Updated by Experiments 38–53, 54A–54Z, 55A–55F, 56D–56H, 56L–56M, 58B.2, v0.8 closeout. Branch: `feature/question-formulation-v0.24`. First-pass reasoning-fidelity v0.8 complete to A–F scope._
### Experiment 58B.2 — Verified Uncertainty Resolution
**Objective:** When the user explicitly verifies a figure and states confidence in its realism, does the engine resolve the existing savings-realism uncertainty rather than merely changing its value?
**Classification: A — UNCERTAINTY CORRECTLY RESOLVED.** One update-only call. The engine correctly resolved `n_savings_realism` (status unknown→resolved), included it in `resolvedUnknownNodeIds`, preserved the £2m figure as "£2,000,000" with verification captured in the reason field ("User verified the projected annual savings including lease exit costs are realistic."). No duplicate uncertainty created. No next question produced — correctly reflecting that no consequential unresolved issue remains. Configured Ollama: qwen-claude:latest at http://192.168.1.111:11434. No production code changed.
---