LEGAL-UQA
收藏arXiv2025-09-30 收录
下载链接：
https://github.com/abdur75648/UTRNet-High-Resolution-Urdu-Text-Recognition
下载链接
链接失效反馈官方服务：
资源简介：
该数据集名为LEGAL-UQA，是首个源自巴基斯坦宪法的乌尔都语法律问答数据集，包含了619组问题与答案，每组都附有相应的法律条文上下文。该数据集的特点是包含了英文和乌尔都语的问题、上下文，以及双语的答案。这些数据是通过光学字符识别（OCR）和人工精炼的方式创建的。数据集的规模为619个问答对，其任务类型为问答。
The dataset named LEGAL-UQA is the first Urdu-language legal question answering (QA) dataset sourced from the Constitution of Pakistan. It contains 619 question-answer pairs, each accompanied by corresponding legal provision context. This dataset features bilingual (English and Urdu) questions, contexts as well as bilingual answers. The data was created through optical character recognition (OCR) and manual refinement, and its task type is question answering.
提供机构：
OpenAI
