让大模型学会根据问题选最佳推理策略，提升解题能力。
Reasoning Strategies in Large Language Models: Can They Follow, Prefer, and Optimize?
- 通过提示工程探索控制模型推理策略的方法
- 单一策略无法稳定提升准确率，但动态选择能优化表现
- 为多策略自适应推理提供新思路，适合复杂逻辑任务研究者
人类推理包含多种策略，适用于不同问题。以往研究发现，大语言模型倾向于依赖单一推理策略，可能限制其在多样化推理挑战中的表现。本文探究提示能否调控大模型的推理策略，并评估其对逻辑问题求解的影响。实验表明，没有单一策略能持续提升准确率，但若模型能自适应选择最优策略，则性能可得到改善。为此，我们提出了引导模型进行策略选择的方法，揭示了提升其推理能力的新途径。
原文摘要 · Abstract (English)
Human reasoning involves different strategies, each suited to specific problems. Prior work shows that large language model (LLMs) tend to favor a single reasoning strategy, potentially limiting their effectiveness in diverse reasoning challenges. In this work, we investigate whether prompting can control LLMs reasoning strategies and assess its impact on logical problem-solving. While our experiments show that no single strategy consistently improves accuracy, performance could be enhanced if models could adaptively choose the optimal strategy. We propose methods to guide LLMs in strategy selection, highlighting new ways to refine their reasoning abilities.
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。