TeleQnA
收藏资源简介:
TeleQnA是由华为技术有限公司巴黎研究中心创建的第一个专门用于评估大型语言模型在电信领域知识的数据集。该数据集包含10,000个问题和答案,内容来源于电信领域的标准和研究文章。数据集的创建过程涉及自动化问题生成框架,并结合人工输入以确保问题质量。TeleQnA主要用于评估如GPT-3.5和GPT-4等大型语言模型在处理电信相关问题时的能力,特别是在理解和应用电信标准方面的表现。此外,该数据集还用于比较专业电信人员与语言模型的表现,以探索语言模型在电信领域的应用潜力。
TeleQnA is the first dataset specifically dedicated to evaluating the knowledge of large language models (LLMs) in the telecommunications domain, developed by the Paris Research Center of Huawei Technologies Co., Ltd. This dataset contains 10,000 question-answer pairs sourced from telecommunications industry standards and academic research articles. The construction of the dataset utilizes an automated question generation framework, supplemented with manual input to guarantee the quality of the questions. TeleQnA is primarily intended to assess the capabilities of LLMs such as GPT-3.5 and GPT-4 when addressing telecommunications-related queries, especially their performance in comprehending and applying telecommunications standards. Furthermore, this dataset can also be used to compare the performance between professional telecommunications practitioners and language models, thereby exploring the application potential of large language models in the telecommunications field.

- 1TeleQnA: A Benchmark Dataset to Assess Large Language Models Telecommunications Knowledge华为技术有限公司巴黎研究中心 · 2023年



