长思维链比多个短思维链推理更高效，尤其在复杂图问题中。
Let Me Think! A Long Chain-of-Thought Can Be Worth Exponentially Many Short Ones
- 用长思维链替代多条短链，提升推理效率。
- 在复杂图连通性任务中，长链性能远超并行短链。
- 适合研究大模型推理机制的学者与工程师。
推理时计算已成为提升大语言模型推理能力的重要扩展方向。然而，尽管表现优异，推理计算的最佳分配方式仍不明确。核心问题是：应优先选择序列扩展（如更长的思维链）还是并行扩展（如多个短思维链的多数投票）？本文通过理论分析与实验验证，在基于图连通性问题的挑战性图分布下，证明了序列扩展相比并行扩展具有指数级优势。实验覆盖多种语言模型，包括从零训练的图连通性模型及大型推理模型，采用不同思维链策略进行对比，充分验证了理论结论。
原文摘要 · Abstract (English)
Inference-time computation has emerged as a promising scaling axis for improving large language model reasoning. However, despite yielding impressive performance, the optimal allocation of inference-time computation remains poorly understood. A central question is whether to prioritize sequential scaling (e.g., longer chains of thought) or parallel scaling (e.g., majority voting across multiple short chains of thought). In this work, we seek to illuminate the landscape of test-time scaling by demonstrating the existence of reasoning settings where sequential scaling offers an exponential advantage over parallel scaling. These settings are based on graph connectivity problems in challenging distributions of graphs. We validate our theoretical findings with comprehensive experiments across a range of language models, including models trained from scratch for graph connectivity with different chain of thought strategies as well as large reasoning models.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。