aisingapore/NLU-Sentiment-Analysis

SEA情感分析数据集用于评估模型识别文本情感极性的能力。该数据集从多个来源采样，包括印尼语、爪哇语、巽他语、泰米尔语、泰语和越南语的情感分析数据。它旨在评估聊天或指令调整的大语言模型（LLM）。SEA Sentiment Analysis is designed to evaluate a models ability to identify the sentiment polarity of a text. It is sampled from sentiment analysis datasets for Indonesian, Javanese, Sundanese, Tamil, Thai, and Vietnamese, and is intended for evaluating chat or instruction-tuned large language models (LLMs).