A stop rule that trusts one score is worse than a dumb budget

Automated excerpt

Three findings. (1) The adaptive policy ties a plain threshold rule ("stop at the first score ≥ 0. 80") everywhere, and it does so by construction: the threshold guard runs first in the guard stack. (2) Both collapse as soon as the judge's score is noisy. A fixed budget of six, which never reads the score, holds 93% at every noise level. (3) A one-line fix, requiring two consecutive scores over the bar, restores 97% true reach at σ = 0. 10 for 1. 4 extra steps. Trusting one score collapses under noise Table 1 is the noise sweep.

Selected automatically from source text; not independently written or fact-checked. Read the original for full context.