arXiv：cs.LG（机器学习，全量分类）· Shivam Gupta·· 14 小时前AI 评分38
验证脉冲与逃离错误共识的代价
Verification Pulses and the Cost of Escaping Wrong Consensus
AI 导读
研究揭示外部验证虽能纠正个体输出，却可能让自强化群体仍困于错误共识。在16个多数更新槽位、14个初始错误的设定下，9次随机检查即可跨过平均场预算阈值，而精确模型需23次才能实现95%最终恢复。一项前瞻性实验记录了13,392条语言模型响应，校准后四种调整方案的比较区间均包含零。
正文
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 $N\log(x_0/b)$ 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.
| Comments: | 18 pages, 5 figures; code and raw experimental records: this https URL |
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.00256 [cs.LG] |
| (or arXiv:2610.00256v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.00256 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Shivam Gupta [view email]
[v1]
Thu, 24 Sep 2026 03:29:27 UTC (147 KB)
来源：arXiv：cs.LG（机器学习，全量分类） · arxiv.org