给推理模型加工具，能真正提升复杂问题解决能力。
Thinking Isn't an Illusion: Overcoming the Limitations of Reasoning Models via Tool Augmentations
- 用代码解释器和草稿板增强推理模型的思考过程。
- 工具辅助下，推理模型在各类难度任务中均优于非推理模型。
- 适合想提升AI逻辑推理能力的研究者和开发者。
大型推理模型（LRMs）是当前大语言模型研究的核心，旨在通过逐步思考过程来应对复杂推理任务。尽管前景广阔，近期实证研究（如苹果公司[Shojaee等, 2025]）表明，这种思考过程可能并未真正提升推理能力，某些情况下无显式推理的LLM反而表现更优。本文重新审视该结论，探究引入工具增强后LRMs的局限性是否依然存在。我们引入两种工具：Python解释器与草稿板，并在苹果公司基准推理谜题上评估三种代表性LLM及其对应的LRM版本。结果显示，在合理使用工具的情况下，LRMs在所有任务复杂度水平上均持续优于其非推理对应模型。这一发现挑战了‘推理是幻觉’的流行观点，凸显了工具增强型推理模型在解决复杂问题上的潜力。
原文摘要 · Abstract (English)
Large Reasoning Models (LRMs) have become a central focus in today's large language model (LLM) research, where models are designed to output a step-by-step thinking process before arriving at a final answer to handle complex reasoning tasks. Despite their promise, recent empirical studies (e.g., [Shojaee et al., 2025] from Apple) suggest that this thinking process may not actually enhance reasoning ability, where LLMs without explicit reasoning actually outperform LRMs on tasks with low or high complexity. In this work, we revisit these findings and investigate whether the limitations of LRMs persist when tool augmentations are introduced. We incorporate two types of tools, Python interpreters and scratchpads, and evaluate three representative LLMs and their LRM counterparts on Apple's benchmark reasoning puzzles. Our results show that, with proper tool use, LRMs consistently outperform their non-reasoning counterparts across all levels of task complexity. These findings challenge the recent narrative that reasoning is an illusion and highlight the potential of tool-augmented LRMs for solving complex problems.
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。