通过正则化优化提示，让大模型推理更省 token 又不丢准确率。
CROP: Token-Efficient Reasoning in Large Language Models via Regularized Prompt Optimization
- 用文本反馈+长度正则化，引导生成简洁推理链
- 在GSM8K等数据集上减少80.6%的token消耗
- 适合需要低成本部署的智能代理系统
使用推理技术的大语言模型虽能提升任务表现，但因生成内容冗长导致延迟高、token开销大。现有自动提示优化（APO）方法只关注准确率，忽略生成长度。本文提出成本正则化提示优化（CROP），在标准准确率反馈外引入文本形式的长度反馈，迫使优化过程生成仅包含关键信息的紧凑推理响应。我们在GSM8K、LogiQA和BIG-Bench Hard等复杂推理数据集上评估该方法，实现80.6%的token消耗降低，性能仅轻微下降。该方法为生产环境中部署高效、低成本的智能体系统提供了实用方案。
原文摘要 · Abstract (English)
Large Language Models utilizing reasoning techniques improve task performance but incur significant latency and token costs due to verbose generation. Existing automatic prompt optimization(APO) frameworks target task accuracy exclusively at the expense of generating long reasoning traces. We propose Cost-Regularized Optimization of Prompts (CROP), an APO method that introduces regularization on response length by generating textual feedback in addition to standard accuracy feedback. This forces the optimization process to produce prompts that elicit concise responses containing only critical information and reasoning. We evaluate our approach on complex reasoning datasets, specifically GSM8K, LogiQA and BIG-Bench Hard. We achieved an 80.6\% reduction in token consumption while maintaining competitive accuracy, seeing only a nominal decline in performance. This presents a pragmatic solution for deploying token-efficient and cost-effective agentic AI systems in production pipelines.
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。