多链图细化与选择用于大语言模型中的可靠推理
Multi-chain Graph Refinement and Selection for Reliable Reasoning in Large Language Models
AI总结：
MGRS通过多链图细化与选择提升大语言模型的推理能力与效率，实现更高准确率和更快速度。
摘要：
大型语言模型（LLMs）复杂的推理能力对其实际应用构成了关键瓶颈。测试时间扩展方法，如树状思维（ToT）和图状思维（GoT），通过引入中间推理结构、树搜索或基于图的探索机制来增强推理能力。然而，它们的推理策略受到多样性有限、冗余搜索分支以及异构推理路径中整合和错误修正不足的限制。为了解决这些限制，我们提出了一种新的推理框架，称为多链图细化与选择（MGRS）。该框架首先为给定问题生成多个多样的推理轨迹，利用复合的自检和互检策略对候选响应进行细化，然后构建推理关系图并估计中间节点的成功率，最后计算累积成功率以选择最可靠的答案和对应的推理轨迹。实验结果表明，MGRS显著推进了推理增强方法的推理能力和计算效率。在六个涵盖四个不同任务的基准数据集上，MGRS实现了82.9%的平均准确率，明显优于最先进的基线方法。值得注意的是，在24点游戏中，MGRS首次实现了100%的准确率，同时相比领先的森林思维框架，速度提升了13.6倍。
英文摘要：
The complex reasoning ability of Large Language Models (LLMs) poses a critical bottleneck for their practical applications. Test-time expansion methods such as Tree-of-Thought (ToT) and Graph-of-Thought (GoT) enhance reasoning by introducing intermediate reasoning structures, tree search, or graph-based exploration mechanisms. However, their reasoning strategies suffer from limited diversity, redundant search branches, and inadequate integration and error correction across heterogeneous reasoning paths. To address these limitations, we propose a novel reasoning framework called Multi-chain Graph Refinement & Selection (MGRS), which first generates multiple diverse reasoning trajectories for a given problem, refines candidate responses using a composite self- and cross-verification strategy, then constructs a reasoning relation graph and estimates the success rate of intermediate nodes, and finally computes cumulative success rates to select the most reliable answer and corresponding reasoning trajectory. Experimental results demonstrate that MGRS significantly advances both the reasoning capability and computational efficiency of reasoning enhancement methods. Across six benchmark datasets spanning four distinct tasks, MGRS achieves an average accuracy of 82.9%, outperforming state-of-the-art baselines by a clear margin of 2.1%. Remarkably, on the 24-point game, MGRS attains 100% accuracy for the first time, while delivering a 13.6x speed-up compared to the leading Forest of Thoughts framework.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
