Verification Pulses and the Cost of Escaping Wrong Consensus

arXiv:2610.00256v1 Announce Type: new Abstract: External verification can correct individual outputs while leaving a self-reinforcing population in the basin of a wrong consensus. We study how the timing and addressing of a fixed verification budget affect recovery in an asynchronous binary register. For a general nonlinear response, we derive the minimum fuel required to cross a basin boundary under a peak verification constraint. For a finite population, an exact birth--death calculation gives the probability of subsequent wrong consensus after a pulse. Our main asymptotic result identifies the critical budget window: a leading term Nlog(x_0/b) and a correction of order sqrt N, with separate variance contributions from repeated verification targets and autonomous amplification after verification stops. The distinction is substantial: with 16 majority-updated slots and 14 initially wrong, 9 random checks cross the mean-field budget threshold, whereas 23 are required for 95% eventual recovery in the exact model. A prospectively specified experiment records 13,392 language-model responses, including calibration and 108 held-out trajectories. Calibration produces different fitted response regimes, but all four adjusted schedule-comparison intervals include zero. A distributional audit also finds that modest mean-prediction error can conceal a large underestimate of terminal consensus occupancy. The results support risk-calibrated reset scheduling under a specified update contract, while explicitly separating it from distinct-target checking and unrestricted evidence broadcast.

Related

• BEHAVE: Functional Behavior Modeling Enables Self-Improving Agents for Hardware Design and Verification
• Efficient Verification of Neural Control Barrier Functions with Smooth Nonlinear Activations
• Fractal basins trap latent reasoning

Source: arXiv cs.LG | 2026-10-02