We present a holistic approach to building a robust and useful natural
language classification system for real-world content moderation. The success
of such a system relies on a chain of carefully designed and executed steps,
including the design of content taxonomies and labeling instructions, data
quality control, an active learning pipeline to capture rare events, and a
variety of methods to make the model robust and to avoid overfitting. Our
moderation system is trained to detect a broad set of categories of undesired
content, including sexual content, hateful content, violence, self-harm, and
harassment. This approach generalizes to a wide range of different content
taxonomies and can be used to create high-quality content classifiers that
outperform off-the-shelf models.

我们提出了一种整体方法来构建一个强大和有用的自然语言分类系统，以进行真实世界的内容审核，该系统的成功依赖于一系列精心设计和执行的步骤，包括内容分类法的设计和标注指令，数据质量控制，捕捉罕见事件的主动学习管道以及各种方法，以使模型具有稳健性并避免过度拟合。我们的审核系统经过训练，可检测许多不良内容的广泛集合，包括色情内容，仇恨内容，暴力，自残和骚扰。这种方法可以概括到各种不同的内容分类法，并可用于创建高质量的内容分类器，胜过现成的模型。