huggingface-KREW/KoCulture-Descriptions

本数据集提供了关于最新韩国语新造词、流行语、迷因及相关文化现象的详细说明，旨在深度提升大型语言模型(LLM)的韩国语理解及生成能力。每个术语的定义、起源、使用方式、文化背景等信息均包含在内。数据集基于从Namuwiki和TrendAward等网站收集的原始资料，使用Claude Sonnet 3.7、Gpt-4o、Gemini 2.5等LLM生成初步说明，并由Hugging Face KREW成员进行审核和精炼。数据集包括标题、内容、日期、类别和来源等字段，并以训练集为主要部分。

This dataset provides detailed descriptions of the latest Korean neologisms, slang, memes, and related cultural phenomena, aiming to deeply enhance the understanding and generation capabilities of large language models (LLMs) in Korean. Each term includes its definition, origin, usage, cultural context, and more. The dataset is based on raw data collected from websites such as Namuwiki and TrendAward, and uses LLMs like Claude Sonnet 3.7, Gpt-4o, Gemini 2.5 to generate initial descriptions, which are then reviewed and refined by Hugging Face KREW members. The dataset includes fields such as title, content, date, category, and source, with the training set being the main focus.