An Empirical Audit of k-NAF Budget Accounting for Anchored Decoding
Abstract
We empirically audit the k-NAF budget-accounting mechanism in Anchored Decoding using (i) a fixed, class-stratified workload (approximately 8,500 randomized executions across six prompt classes) and (ii) an adaptive prompt-search procedure targeting high proxy spend ratios. On the fixed workload, mean cumulative KL spend remains far below the sequence-level budgets K in {600, 1000}, and an empirical Bernstein-style proxy stays below K for every class; surface-overlap diagnostics (ROUGE-L and 5-gram Jaccard) are correspondingly small. Adaptive search increases the proxy spend ratio but does not produce clear budget exhaustion. On a held-out copyright-domain workload at k = 3, several prompts exhibit proxy ratios above 1 under early-stopped evaluations with small realized sample sizes; re-evaluating the same prompts with larger allocation reduces the proxy ratio to the range [0.26, 0.40] under comparable mean spend, consistent with proxy artifacts rather than per-trajectory budget failures.
Cite
@article{arxiv.2605.28001,
title = {An Empirical Audit of k-NAF Budget Accounting for Anchored Decoding},
author = {J. Vijayavallabh},
journal= {arXiv preprint arXiv:2605.28001},
year = {2026}
}
Comments
19 pages, 4 figures, 9 main pages remaining supplementary and appendix