融合双向思维链与奖励机制：一种增强大语言模型中国非物质文化遗产问答能力的方法
Fusing Bidirectional Chains of Thought and Reward Mechanisms A Method for Enhancing Question-Answering Capabilities of Large Language Models for Chinese Intangible Cultural Heritage
AI总结：
针对非遗领域大模型微调中的偏差、错误继承与遗忘问题，提出在 ICH-Qwen 上融合双向思维链和加权奖励机制的训练方法，并验证其问答性能、消融效果与跨领域泛化能力。
摘要：
大语言模型（LLM）的快速发展为特定领域大语言模型的进步提供了重要支撑和机遇。然而，使用非物质文化遗产（ICH）数据微调这些大模型不可避免地面临偏差、知识继承错误以及灾难性遗忘等挑战。为解决这些问题，我们提出了一种融合双向思维链与奖励机制的新型训练方法。该方法建立在 ICH-Qwen 之上，后者是一个专门面向非物质文化遗产领域设计的大语言模型。所提方法使模型不仅能够进行正向推理，还通过利用反向提问和反向推理激活模型的潜在知识，提高生成答案的准确性。此外，训练过程中引入奖励机制以优化决策过程。该机制通过采用不同权重方案的结构评估和内容评估，提升模型输出质量。我们在 ICH-Qwen 上开展了比较实验，结果表明，在问答任务的准确率、Bleu-4 和 Rouge-L 分数方面，我们的方法优于零样本、逐步推理、知识蒸馏和问题增强方法。此外，本文通过消融实验强调了双向思维链与奖励机制相结合的有效性。同时，还开展了一系列泛化性实验，结果显示所提方法在金融、Wikidata 和 StrategyQA 等领域的多种特定领域数据集和先进模型上均带来提升。这表明该方法能够适应多个领域，并为未来在不同领域应用中的模型训练提供了一种有价值的途径。
英文摘要：
The rapid development of large language models (LLMs) has provided significant support and opportunities for the advancement of domain-specific LLMs. However, fine-tuning these large models using Intangible Cultural Heritage (ICH) data inevitably faces challenges such as bias, incorrect knowledge inheritance, and catastrophic forgetting. To address these issues, we propose a novel training method that integrates a bidirectional chains of thought and a reward mechanism. This method is built upon ICH-Qwen, a large language model specifically designed for the field of intangible cultural heritage. The proposed method enables the model to not only perform forward reasoning but also enhances the accuracy of the generated answers by utilizing reverse questioning and reverse reasoning to activate the model's latent knowledge. Additionally, a reward mechanism is introduced during training to optimize the decision-making process. This mechanism improves the quality of the model's outputs through structural and content evaluations with different weighting schemes. We conduct comparative experiments on ICH-Qwen, with results demonstrating that our method outperforms 0-shot, step-by-step reasoning, knowledge distillation, and question augmentation methods in terms of accuracy, Bleu-4, and Rouge-L scores on the question-answering task. Furthermore, the paper highlights the effectiveness of combining the bidirectional chains of thought and reward mechanism through ablation experiments. In addition, a series of generalizability experiments are conducted, with results showing that the proposed method yields improvements on various domain-specific datasets and advanced models in areas such as Finance, Wikidata, and StrategyQA. This demonstrates that the method is adaptable to multiple domains and provides a valuable approach for model training in future applications across diverse fields.
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。
