用进化算法优化推理模型，让回答更短更准
Multi-objective Evolutionary Merging Enables Efficient Reasoning Models
- 设计多目标进化框架，自动平衡准确率与输出长度
- 在6个数学题集上压缩50%以上文本量，14B模型压缩潜力更大
- 适合需要高效推理的AI应用，如实时问答系统
推理模型通过长思维链实现复杂问题求解，但带来高昂推理开销。长转短（L2S）问题旨在保持精度的同时减少生成文本量。现有无训练合并方法依赖固定超参的简单加权，难以兼顾性能。本文提出Evo-L2S，一种多目标进化模型合并框架，显式优化准确率与输出长度，恢复帕累托前沿。为降低计算成本，引入基于熵的子集采样技术，显著减少适应度评估开销。在六个数学推理基准上，Evo-L2S在1.5B和7B规模下将推理长度减少超50%，同时保持或提升准确率；在14B规模下，更陡峭的帕累托前沿表明可压缩性取决于模型内在可压缩性。结果表明，推理模型可在不牺牲性能前提下大幅精简。
原文摘要 · Abstract (English)
Reasoning models achieve strong performance on complex problems by leveraging long chains of thought, but this deliberate reasoning incurs substantial inference-time cost. The Long-to-Short (L2S) reasoning problem seeks to preserve accuracy while reducing generated tokens. Yet, current training-free model merging approaches rely on brittle, fixed-hyperparameter arithmetic methods that force suboptimal compromises. We introduce \textbf{Evo-L2S}, a multi-objective evolutionary model merging framework that explicitly optimizes accuracy and output length to recover a Pareto front of merged models. To make this search computationally tractable, we propose an entropy-based subset sampling technique that substantially reduces fitness-estimation overhead. Across six mathematical reasoning benchmarks, Evo-L2S reduces reasoning length by over 50\% at the 1.5B and 7B scales while preserving or improving problem-solving accuracy; at 14B, the steeper Pareto front reveals that attainable compression depends on the intrinsic compressibility of the reasoning model. Overall, Evo-L2S shows that reasoning models can be made substantially more concise while preserving strong problem-solving performance.
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。