Jackrong/financial-economics-reasoning

金融经济推理数据集是通过高级推理蒸馏技术构建的，使用了qwen-3-235b-a22b-thinking-2507教师模型处理开源的BAAI/IndustryInstruction_Finance-Economics数据集，该数据集包含122,378条金融、经济和商业领域的双语文（中英）条目。与标准蒸馏数据集仅提供最终答案不同，此数据集保留了从教师模型蒸馏出的完整推理链（CoT）。适用于金融、经济和商业领域的对话模型训练或微调，研究推理蒸馏和大型语言模型的解释性，以及提高问答、总结和推理密集型任务的下游性能。

The financial-economics-reasoning dataset is constructed using advanced Inference Distillation techniques from the BAAI/IndustryInstruction_Finance-Economics dataset, containing 122,378 bilingual (Chinese-English) entries in the fields of finance, economics, and business. Unlike standard distillation datasets that only provide final answers, this dataset retains the full reasoning chains (CoT) distilled from the Teacher Model. It is suitable for training or fine-tuning dialogue models in the fields of finance, economics, and business, researching reasoning distillation and interpretability in large language models, and improving downstream performance in tasks such as Q&A, summarization, and reasoning-intensive applications.