网络百科全书的审查：对NLP模型的启示
Censorship of Online Encyclopedias: Implications for NLP Models
摘要：
虽然人工智能为世界各地人们使用的许多工具提供了支柱，但最近的工作引起了人们的关注，即为人工智能提供动力的算法并非没有政治，刻板印象和偏见。虽然这一领域的大多数工作都集中在人工智能加剧现有不平等和歧视的方式上，但很少有工作研究政府如何积极塑造训练数据。我们描述了审查如何影响维基百科语料库的发展，这些语料库经常用于NLP算法的预训练输入。我们发现，在百度百科（一个在线中文百科全书）上训练的词嵌入，与经常被屏蔽但未经审查的中文维基百科相比，在形容词和一系列关于民主、自由、集体行动、平等、人民和历史事件的概念之间有着非常不同的关联。我们通过研究这些差异在下游人工智能应用中的使用来研究它们的影响。我们的论文展示了政府的压制、审查和自我审查如何影响训练数据和从中提取的应用程序。
英文摘要：
While artificial intelligence provides the backbone for many tools people use around the world, recent work has brought to attention that the algorithms powering AI are not free of politics, stereotypes, and bias. While most work in this area has focused on the ways in which AI can exacerbate existing inequalities and discrimination, very little work has studied how governments actively shape training data. We describe how censorship has affected the development of Wikipedia corpuses, text data which are regularly used for pre-trained inputs into NLP algorithms. We show that word embeddings trained on Baidu Baike, an online Chinese encyclopedia, have very different associations between adjectives and a range of concepts about democracy, freedom, collective action, equality, and people and historical events in China than its regularly blocked but uncensored counterpart - Chinese language Wikipedia. We examine the implications of these discrepancies by studying their use in downstream AI applications. Our paper shows how government repression, censorship, and self-censorship may impact training data and the applications that draw from them.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
