让大模型自动判断何时停止推理，减少冗余步骤
OS-Pruner: Pruning Chains-of-Thought of Reasoning Models via Optimal Stopping
- 将推理链修剪建模为最优停止问题，动态判断终止时机
- 在多个基准上实现20%-60%生成长度压缩，准确率几乎不变
- 无需重训练，轻量插件式设计，适合追求效率的部署场景
大型语言模型通过思维链（Chain-of-Thought, CoT）提示在复杂推理任务中表现优异。然而，这些模型常出现‘过度计算’现象，生成冗余推理步骤，增加延迟和成本却未提升准确率。尽管已有研究证明可显著压缩CoT轨迹，但现有方法多依赖固定思考预算、启发式过滤、次优的早期退出机制或昂贵的再训练。本文提出OS-Pruner，一种轻量级插件式框架，将思维链剪枝建模为最优停止问题。给定推理前缀，OS-Pruner通过优化显式效用函数（权衡最终答案准确率与生成长度）来学习是否值得继续推理。该新范式使模型能动态评估推理链的充分终止点。OS-Pruner在训练与推理阶段均轻量高效，支持用户对推理投入与准确率之间的权衡进行细粒度控制。在多种推理基准与基础模型上，其生成长度减少20%-60%，准确率损失极小。
原文摘要 · Abstract (English)
Large Language Models (LLMs) have achieved remarkable success in complex reasoning tasks through Chain-of-Thought (CoT) prompting. However, these models often exhibit "computational overthinking," generating redundant reasoning steps that increase latency and cost without improving accuracy. Recent studies suggest that CoT trajectories can be significantly pruned, yet existing methods often rely on forcing a static thinking budget, heuristic filtering, sub-optimal early exit via classification, or expensive re-training. In this paper, we introduce OS-Pruner, a lightweight plug-in framework that formulates chain-of-thought pruning as an optimal stopping problem. Given a reasoning prefix, OS-Pruner learns whether further reasoning is worth its token cost by optimizing an explicit utility that trades off final-answer accuracy against generated length. Our novel formulation enables the model to dynamically assess the sufficient point of termination for a reasoning chain. OS-Pruner is designed to be lightweight during both training and inference, and to provide users with fine-grained control over the reasoning-effort vs. accuracy trade-off. On diverse reasoning benchmarks and base models, OS-Pruner achieves 20-60\% reduction in generation length with minimal accuracy sacrifice.
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。