avemio/German-RAG-SFT-Alpaca-HESSIAN-AI

German-RAG-SFT数据集是一个基于德语维基百科构建的监督微调任务数据集，专为增强语言模型的RAG（检索增强生成）能力而设计。数据集包括多个任务配置，如分类、提取、OCR校正、带时间差异或不带时间差异的问题回答等，每个配置都有相应的训练和测试数据文件。数据集适用于多种NLP任务，如文本分类、问题回答、摘要生成等，并提供了丰富的任务示例。

The German-RAG-SFT dataset is a supervised fine-tuning task dataset built on the German Wikipedia, specifically designed to enhance the RAG (Retrieval Augmented Generation) capabilities of language models. The dataset includes multiple task configurations such as classification, extraction, OCR correction, question answering with or without time difference, and each configuration has corresponding training and test data files. The dataset is suitable for various NLP tasks such as text classification, question answering, summarization, and provides a wealth of task examples.