Evidence record · 2026
The MAVS-GC benchmark program
The complete verified record behind MAVS-GC: clean accuracy, robustness under corruption, stability, dynamic sequential validation, and the Diagnostic Sciences correlated-failure fix.
Clean accuracy
Chapter 10A prevents overclaiming. Under clean benchmark conditions, MAVS-GC is competitive but not dominant: it produced positive metric deltas in 79 of 288 comparisons, improved accuracy over Veto MAVS in 2 of 8 and over the Static Weighted ensemble in 0 of 8. Governance shifts the error profile — often raising precision and reducing false positives while lowering recall/F1 — rather than raising the accuracy ceiling.
Robustness under corruption
Chapter 10B is the strongest classical signal: under stress, MAVS-GC fails more safely. Across four datasets and nine corruption families, governed consensus suppresses unsafe acceptance by up to ~200×.
Chapter 10B · accuracy vs. unsafe acceptance
Under specialist-failure corruption, Pure MAVS-GC keeps accuracy high while unsafe acceptance stays near zero — roughly 20× lower than ensemble baselines and 34× lower than a single model.
Reproducibility & stability
Clean-condition reproducibility gains are limited, but stability preservation strengthens as corruption increases.
| Metric | Pure MAVS-GC | Baseline |
|---|---|---|
| Prediction stability | 0.971615 | 0.952713 |
| Decision stability | 0.975770 | 0.958762 |
| Consensus stability | 0.979332 | 0.963946 |
| Trace stability | 0.967976 | 0.959693 |
Dynamic validation
Chapter 10D moves from static rows to sequential episodes with corruption schedules, recovery periods, and hidden safety labels. The full minimum run produced 383,200 trace records across 13 experiments (E1–E5), 2,530 episodes, and 12,161 failure cards, with trace and audit-trace completeness at 1.0000.
The finding is precise: MAVS-GC achieved near-zero unsafe acceptance across the reported rows, but behaved conservatively — sometimes paying for safety with elevated false rejection. It is auditable and safe, not universally superior.
| Environment | Reward | UAR | FRR | Rank |
|---|---|---|---|---|
| Text Safety Stream | 0.9932 | 0.0000 | 0.0074 | 2 / 14 |
| Tool-Use Security | 0.8974 | 0.0000 | 0.1703 | 4 / 14 |
| Synthetic Ops | 0.8578 | 0.0000 | 0.1804 | 5 / 14 |
Negative result — motivates DS-CF
On correlated representation collapse, MAVS-GC avoided unsafe acceptance (UAR 0.0000) but collapsed into rejection: FRR 1.0000, mean reward 0.0250, collapse sensitivity −1.0000. Chapter 10D explicitly does not claim MAVS-GC solves correlated failure — which is exactly the weakness the next result fixes.
See how Diagnostic Sciences (DS-CF) fixes this ↓Specials-1 · Diagnostic Sciences (DS-CF)
The Diagnostic Sciences correlated-failure fix retargets governance from punishing correlation to punishing harmful correlation. It is a governance-only change (no training) that removes the false-rejection collapse from Chapter 10D while preserving zero unsafe acceptance.
Phase 4 A/B · governance-only, no model training · traces 1.0000 complete
DS-CF changes governance from punishing correlation to distinguishing harmful correlation from safe consistency. It removes the false-rejection failure mode while keeping unsafe acceptance at zero — validated across 1,794 governance decisions.
Phase 5 · original MAVS-GC vs. DS-CF · all DS-CF unsafe acceptance = 0
Average false rejection drops from 0.182 to 0.009 across eight families with zero unsafe acceptance. The only residual is multi-agent triage, where safe-consistency evidence was masked — an evidence-availability limit, not a hard-veto failure.
A trace audit over 1,794 governance decisions found zero raw-correlation-only vetoes, 94 valid conjunctive hard vetoes, zero hard-veto rule violations, and 40 / 40 ambiguous cases escalated correctly.
Governance ablations
Chapter 11B decomposes the robustness signal by removing one governance mechanism at a time. Trace persistence, diagnostics, and the severity→threshold chain explain most of the degradation.