aida-ugent/llm-censorship

这个数据集用于衡量大型语言模型（LLM）中的软审查（信息的选择性省略）现象。它包含了14个来自不同地区（西方国家、中国和俄罗斯）的最先进LLM在所有六个官方联合国语言中关于政治人物的回应。该数据集旨在提供洞见，了解LLM在讨论政治话题时如何在何时拒绝提供信息或选择性省略细节，目标是提高对LLM审查实践中意识形态偏见的透明度。

This dataset measures soft censorship (selective omission of information) in large language models (LLMs). It contains responses from 14 state-of-the-art LLMs from different regions (Western countries, China, and Russia) when prompted about political figures in all six official UN languages. The dataset is designed to provide insights into how and when LLMs refuse to provide information or selectively omit details when discussing political topics, with the goal of enhancing transparency regarding ideological biases in LLM moderation practices.