arXiv cs.CL· Amruta Parulekar, Jinu Lee, Dilek Hakkani-T\"ur, Hari Sundaram·· 11 小时前AI 评分35
Reasoning Consensus：通过加权 DAG 聚合实现 LLM 推理的结构化集成
Reasoning Consensus: Structural Ensembling of LLM Reasoning via Weighted DAG Aggregation
AI 导读
Reasoning Consensus（RC）从每条 CoT 中提取带类型的推理图，跨模型合并等价步骤并按支持链数量加权，从而暴露每一步的支撑程度。在覆盖法律、科学、医学、叙事推理与逻辑的 7 个基准上，RC 准确率达 69.4%，与多数投票和辩论的 69.0%-70.3% 相当。用户研究中参与者阅读 RC 图快于辩论记录或投票统计，18 人中有 14 人偏好未来使用该方式。
正文 · 原文
Abstract:Large Language Models (LLMs) reason in free-form chain-of-thought (CoT) without revealing which steps were well supported and which alternatives were rejected. In high-stakes domains, users then cannot judge if an answer rests on sound reasoning. Majority voting and multi-agent debate improve answers using model agreement, but do not expose agreement on individual steps. Reasoning Consensus (RC) extracts a typed reasoning graph from each CoT, merges equivalent steps across graphs from diverse models, and weights every step by how many chains attest to it. RC reveals this step-wise support and the varied justifications behind each answer. On 7 benchmarks across law, science, medicine, narrative reasoning and logic, RC matches majority voting and debate in accuracy (69.4% vs. 69.0-70.3%). An LLM judge prefers RC's best-supported graph to other graphs for the same answer. In a user study, participants read RC graphs faster than debate transcripts or vote tallies, and 14 of 18 preferred them for future use.
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2607.27783 [cs.CL] |
| (or arXiv:2607.27783v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2607.27783 arXiv-issued DOI via DataCite |
Submission history
From: Amruta Parulekar [view email]
[v1]
Thu, 30 Jul 2026 07:14:50 UTC (819 KB)
[v2]
Fri, 30 Oct 2026 20:21:54 UTC (446 KB)
来源：arXiv cs.CL · arxiv.org