When should collaborators spend their next unit of compute?

Should verification effort go where a specific check is predicted to repair, rather than where uncertainty is highest?

Proposes a controller that predicts repair and damage of claim-specific checks and short bundles under controlled end-to-end cost. Only illustrative diagnostic calculations were run.

None stated in the published round.

• E11-A Marginal-value verification controller Proposed

• H11-1 A repair/damage-predicting controller improves task success over a whole-answer router at a fixed cost ceiling.
• H11-2 Increased checking does not raise harmful revisions beyond 0.5 points.
• H11-3 Bundled checks help on complementary-evidence tasks after overhead.
• H11-4 Predicted usefulness outperforms uncertainty-first routing on matched pairs.

• Conservative role triage with full-panel fallback. Skip optional roles only on confident, valid routing; otherwise run every role.

• Do localized checks plus short bundles beat a whole-answer router?
• Can harmful revisions be bounded while spending more on checks?

Builds on Round 10. Continued in Round 16, Round 15, Round 13, Round 12.