Learning to Be Wrong: Task-Sufficient Inexact Computation

Neural efficiency methods usually optimize how much computation to spend. We study how much local exactness is worth paying for. A cheaper realization may violate a conventional local contract while preserving downstream utility. We evaluate each candidate with measured runtime against a matched exact reference, a mechanism-specific local-fidelity diagnostic, and signed downstream task regret. Across precision-relaxed, incomplete, and contract-relaxed realizations, nominal work, local fidelity, hardware cost, and task utility frequently disagree. Our main case study groups several target positions in ESM-2 masked-marginal scoring so that one forward evaluates multiple mutations under a changed conditioning context. At $g=4$, the same frozen random-group procedure yields 3.93–3.99$\times$ measured scoring speedup with mean signed DMS regret 0.0010–0.0012 on TEM-1, influenza HA, and a DHFR assay held out from policy development; at $g=16$, speedup reaches 12.34 - 15.59$\times$ with assay-dependent regret. Pairwise interference is predictable in development, while held-out DHFR shows that a fidelity-aware grouping rule can improve exact-score agreement and still worsen task regret. These results make task regret the governing quantity for inexact execution: spend exactness where it changes the task.