Before the Answer Closes: Detecting Missing Decision Conditions

At a fixed total inference and source-access budget, does requiring a validated pair of evidence-compatible scenarios with different decisions reduce unsupported commitment more effectively than a strong missing-information checklist, without reducing useful completion or increasing harmful revision?

Proposes a bounded contrastive sufficiency probe and a separate source-coverage probe for missing decision conditions. Executed finite-world diagnostics cover 6,912 Boolean function-observation pairs and 37 named checks, with twelve developer fixtures. E16-B and H16-1/H16-2/H16-3 are proposed and untested with language models. No new Encevra language-model benchmark, deployment, production change, or calibrated release guarantee.

• A supported fact does not establish decision sufficiency: the exact fictional policy counterexample leaves both review decisions possible.
• Exact finite-world enumeration yields 3,778 determined and 3,134 underdetermined pairs among 6,912 pairs. These are checker diagnostics, not model performance.
• A valid decision-changing pair establishes a gap within the modeled scope; failed bounded search does not certify completeness. Scope expansion can reopen a decision.

• E16-A Exact finite-world sufficiency diagnostics Executed (diagnostic)
• E16-B Bounded missing-condition study Proposed

• H16-1 Validated alternatives improve supported task completion and reduce unsupported commitment relative to a strong checklist at the same budget.
• H16-2 Scope discovery helps more when a relevant rule is absent from the representation than when only a named value is missing.
• H16-3 Decision relevance preserves useful completion on sufficient and irrelevant-unknown controls while reducing unsupported closure.

Proposed mechanism only. No Noetis behavior, release gate, risk value, routing, or production change.

• Does validated alternative generation improve supported task completion over a strong checklist at matched budget?
• Does scope discovery help when a relevant rule is absent from the representation?
• Can decision relevance reduce unsupported closure without needless questions or reduced useful completion?

Builds on Round 15, Round 14, Round 13, Round 12, Round 11, Round 10, Round 9. Continued in Round 19, Round 18, Round 17.