通过知识注入实现高效学习内容检索
Efficient Learning Content Retrieval with Knowledge Injection
AI总结：
针对在线教育资源检索困难，本文基于 QLoRA 微调 Phi-2/Phi-3，并结合由多格式文档构建的 RAG 知识库，通过多指标评估选出适合实际部署的低资源领域聊天机器人方案。
摘要：
随着在线教育平台的兴起，各个领域的教育内容日益丰富。要在大量可用资源中导航并找到最合适的培训内容可能十分困难，尤其是在信息与通信技术（ICT）等包含许多相互关联领域的方向。本研究提出了一种资源需求有限的特定领域聊天机器人应用，利用多个版本的 Phi 语言模型帮助学习者获取教育内容。在所提出的方法中，Phi-2 和 Phi-3 模型使用 QLoRA 进行微调。微调所需数据来自 Huawei Talent Platform，该平台提供计算机科学领域不同专业水平的课程。研究使用 RAG 系统为模型提供支持，模型由 500 个问答对进行微调。此外，还从 JSON、PPT 和 DOC 等不同格式中提取了共计 420 个内容问答对，以构建供 RAG 系统使用的向量数据库。通过将微调模型与 RAG 方法结合使用，获得了具备不同能力的聊天机器人。向生成的聊天机器人提出的问题及其回答被分别保存，并使用 ROUGE、BERTScore、METEOR 和 BLEU 指标进行评估。由 RAG 支持的 Phi-2 模型精确率为 0.84，F1 分数为 0.82。除了 4 个不同类别中的共计 13 项不同评估指标外，研究还将每个模型的回答与所创建内容进行比较，并为实际应用选择了最合适的方法。
英文摘要：
With the rise of online education platforms, there is a growing abundance of educational content across various domain. It can be difficult to navigate the numerous available resources to find the most suitable training, especially in domains that include many interconnected areas, such as ICT. In this study, we propose a domain-specific chatbot application that requires limited resources, utilizing versions of the Phi language model to help learners with educational content. In the proposed method, Phi-2 and Phi-3 models were fine-tuned using QLoRA. The data required for fine-tuning was obtained from the Huawei Talent Platform, where courses are available at different levels of expertise in the field of computer science. RAG system was used to support the model, which was fine-tuned by 500 Q&A pairs. Additionally, a total of 420 Q&A pairs of content were extracted from different formats such as JSON, PPT, and DOC to create a vector database to be used in the RAG system. By using the fine-tuned model and RAG approach together, chatbots with different competencies were obtained. The questions and answers asked to the generated chatbots were saved separately and evaluated using ROUGE, BERTScore, METEOR, and BLEU metrics. The precision value of the Phi-2 model supported by RAG was 0.84 and the F1 score was 0.82. In addition to a total of 13 different evaluation metrics in 4 different categories, the answers of each model were compared with the created content and the most appropriate method was selected for real-life applications.
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。
