CoT-Evo：面向科学推理的思维链进化蒸馏
CoT-Evo: Evolutionary Distillation of Chain-of-Thought for Scientific Reasoning
AI总结：
CoT-Evo提出进化式思维链蒸馏框架，通过多样化轨迹池、知识增强和迭代优化，生成高质量科学推理数据，微调紧凑模型达到最先进性能。
摘要：
虽然从先进的大语言模型（LLMs）中进行思维链（CoT）蒸馏在通用推理任务中已被证明有效，但在科学领域，由于高复杂性和专业知识要求，即使是先进的模型也常常产生不正确或肤浅的推理，因此该方法在科学领域面临困境。直接从此类有缺陷的输出中进行蒸馏会导致训练数据质量低下，并限制较小学生模型的性能。为克服这一问题，我们提出了CoT-Evo，一种进化式思维链蒸馏框架。它首先从多个LLM思考者构建一个多样化的推理轨迹池，通过自动检索的领域知识丰富这些轨迹，并使用新颖性驱动的选择、反思性重组和变异迭代地改进轨迹。改进过程由适应度函数指导，该函数评估答案正确性、连贯性和知识利用的有效性。这产生了一个针对科学推理定制的高质量CoT数据集。我们利用这个进化后的数据集微调一个紧凑模型，该模型在科学推理基准上取得了最先进的性能。我们的工作建立了一种从多样且易出错的LLMs中合成高保真科学推理数据的可扩展方法。
英文摘要：
While chain-of-thought (CoT) distillation from advanced large language models (LLMs) has proven effective in general reasoning tasks, it struggles in scientific domains where even advanced models often produce incorrect or superficial reasoning due to high complexity and specialized knowledge requirements. Directly distilling from such flawed outputs results in low-quality training data and limits the performance of smaller student models. To overcome this, we propose CoT-Evo, an evolutionary CoT distillation framework. It begins by constructing a diverse pool of reasoning trajectories from multiple LLM thinkers, enriches them with automatically retrieved domain knowledge, and iteratively refines the trajectories using novelty-driven selection, reflective recombination and mutation. The refinement is guided by a fitness function that evaluates answer correctness, coherence, and effective knowledge utilization. This results in a high-quality CoT dataset tailored for scientific reasoning. We employ this evolved dataset to fine-tune a compact model, which achieves state-of-the-art performance on scientific reasoning benchmarks. Our work establishes a scalable approach to synthesizing high-fidelity scientific reasoning data from diverse and fallible LLMs.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
