通过多轮精炼压缩思维链，提升推理速度与准确率。
From Long to Lean: Performance-aware and Adaptive Chain-of-Thought Compression via Multi-round Refinement
- 利用令牌弹性现象，多轮迭代动态压缩思维链。
- 平均减少47个令牌，准确率提升5.6个百分点。
- 可预测测试性能，适合高效模型选型与部署。
思维链（CoT）推理虽能提升复杂任务表现，但因冗长导致显著延迟。我们提出多轮自适应思维链压缩框架（MACC），利用令牌弹性现象——过小的令牌预算反而增加输出长度——通过多轮精炼逐步压缩CoT。该自适应策略可为每个输入确定最优压缩深度。实验显示，相比最先进基线，方法平均准确率提升5.6%，平均减少47个令牌，并显著降低延迟。此外，我们证明测试时性能（准确率与令牌长度）可基于训练集中的可解释特征（如困惑度、压缩率）可靠预测。在不同模型上验证表明，该方法无需重复微调即可实现高效模型选择与性能预估，证实了CoT压缩既有效又可预测。代码将发布于 https://github.com/Leon221220/MACC。
原文摘要 · Abstract (English)
Chain-of-Thought (CoT) reasoning improves performance on complex tasks but introduces significant inference latency due to verbosity. We propose Multiround Adaptive Chain-of-Thought Compression (MACC), a framework that leverages the token elasticity phenomenon--where overly small token budgets can paradoxically increase output length--to progressively compress CoTs via multiround refinement. This adaptive strategy allows MACC to determine the optimal compression depth for each input. Our method achieves an average accuracy improvement of 5.6 percent over state-of-the-art baselines, while also reducing CoT length by an average of 47 tokens and significantly lowering latency. Furthermore, we show that test-time performance--accuracy and token length--can be reliably predicted using interpretable features like perplexity and compression rate on the training set. Evaluated across different models, our method enables efficient model selection and forecasting without repeated fine-tuning, demonstrating that CoT compression is both effective and predictable. Our code will be released in https://github.com/Leon221220/MACC.
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。