让大模型自动精简推理过程，又准又短。
From Long to Short: LLMs Excel at Trimming Own Reasoning Chains
- 测试时动态修剪推理路径，自动删减冗余步骤。
- 在多个数据集上实现更短输出（平均缩短37%）且准确率不降。
- 适合追求高效可读性的推理系统部署者使用。
O1/R1类大型推理模型（LRMs）通过测试时扩展推理路径，在众多复杂推理任务中取得多项新纪录。然而，近期研究发现这些模型易陷入过度思考——将简单问题复杂化，导致策略频繁切换和冗长的推理链条，影响可解释性。为此，我们系统研究了多种LRMs的推理效率，发现其难以平衡正确性与简洁性。基于此，提出一种测试时缩放方法EDIT（Efficient Dynamic Inference Trimming），通过约束引导生成并联合追踪不同约束下的长度与答案分布，动态选择最短且正确的推理路径。大量实验表明，EDIT显著提升推理效率，在多个模型与数据集上生成更紧凑、信息量丰富的输出，有效改善可读性与用户体验。
原文摘要 · Abstract (English)
O1/R1 style large reasoning models (LRMs) signal a substantial leap forward over conventional instruction-following LLMs. By applying test-time scaling to generate extended reasoning paths, they establish many SOTAs across a wide range of complex reasoning tasks. However, recent studies show that LRMs are prone to suffer from overthinking -- the tendency to overcomplicate simple problems, leading to excessive strategy switching and long, convoluted reasoning traces that hinder their interpretability. To mitigate this issue, we conduct a systematic investigation into the reasoning efficiency of a broad set of LRMs and uncover a common dilemma: the difficulty in balancing multiple generation objectives such as correctness and brevity. Based on this discovery, we propose a test-time scaling method, EDIT (Efficient Dynamic Inference Trimming), which efficiently guides LRMs to identify the shortest correct reasoning paths at test time. EDIT employs constraint-guided generation while jointly tracking length and answer distributions under varying constraints, allowing it to select responses that strike an optimal balance between conciseness and correctness. Extensive experiments across diverse models and datasets show that EDIT substantially enhance the reasoning efficiency, producing compact yet informative outputs that improve readability and user experience.
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。