QCRI/LlamaLens-Hindi

LlamaLens是一个专门用于分析新闻和社交媒体内容的多语言大型语言模型（LLM）。它支持阿拉伯语、英语和印地语，涵盖了19种自然语言处理（NLP）任务，并利用了52个数据集。数据集包括情感分析、仇恨言论检测、虚假新闻检测等多种任务，每个任务都有详细的训练、测试和开发集数据。文件格式为JSONL，包含输入文本、输出标签、任务类型、语言等信息。

LlamaLens is a specialized multilingual large language model (LLM) designed for analyzing news and social media content. It supports Arabic, English, and Hindi, covering 19 natural language processing (NLP) tasks and leveraging 52 datasets. The dataset includes tasks such as sentiment analysis, hate speech detection, fake news detection, and more, with detailed training, testing, and development set data for each task. The file format is JSONL, containing input text, output labels, task types, language information, and more.