arXiv cs.CY· Mila Gorecki, Moritz Hardt·· 11 小时前AI 评分34
推理会让语言模型更一致吗？arXiv 研究区分“推理干预”与“推理行为”
Does Reasoning Make Language Models Agree?
AI 导读
一项 arXiv 研究区分“推理作为干预”与“推理作为可观察行为”，在来自两家开发者的九个模型和三个分类任务上发现：开启推理会提高模型对同一个体的预测一致性，但推理链更长的个体反而获得更不一致的跨模型预测，两种模式在控制准确率差异后依然成立。
正文 · 原文
Abstract:Language models face two opposing concerns: monoculture, in which models err in correlated ways, and multiplicity, in which equally accurate models make conflicting decisions. The rise of reasoning models has complicated this debate: On the one hand, reasoning may increase model agreement if models anchor on the same salient features or follow similar reasoning paths. On the other hand, stochastic reasoning traces may introduce additional branch points leading to more diverse predictions. We resolve this tension by distinguishing reasoning as an intervention from reasoning as an observed behavior. We compare agreement under controlled changes in reasoning effort with agreement across observed reasoning lengths, measuring both against the agreement expected from model accuracy. Across nine models from two developers and three classification tasks, enabling reasoning increases agreement between models evaluated on the same individuals. Yet the individuals who elicit longer reasoning traces receive less consistent predictions across models. Both patterns persist after accounting for differences in accuracy. Asking models to reason more thus promotes agreement, while observing them reason longer signals disagreement.
| Subjects: | Machine Learning (cs.LG); Computation and Language (cs.CL); Computers and Society (cs.CY) |
| Cite as: | arXiv:2610.41691 [cs.LG] |
| (or arXiv:2610.41691v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.41691 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Mila Gorecki [view email]
[v1]
Fri, 30 Oct 2026 12:57:53 UTC (1,445 KB)
来源：arXiv cs.CY · arxiv.org