论文标题
信息性和不变性：关于自然语言虚假相关性的两个观点
Informativeness and Invariance: Two Perspectives on Spurious Correlations in Natural Language
论文作者
论文摘要
虚假的相关性是对自然语言处理系统的可信赖性的威胁，激励研究识别和消除它们的方法。但是，解决虚假相关性问题需要更加清楚它们在语言数据中的出现。 Gardner等人（2021年）认为，语言的组成性质意味着\ emph {All}标签和单个“输入特征”之间的相关性是虚假的。本文在玩具示例的背景下分析了该提案，展示了三种不同的条件，这些条件可以引起简单的PCFG中特征标签相关性。将玩具示例链接到结构化因果模型表明，即使标签对特征的干预是不变的，也可能会出现（1）特征标签的相关性，并且（2）即使标签对特征上的干预措施敏感，也可能缺少特征标签标签的相关性。由于输入功能将与在非常罕见的情况下的所有标签中单独相关，因此必须应用域知识来识别构成真正鲁棒性威胁的虚假相关性。
Spurious correlations are a threat to the trustworthiness of natural language processing systems, motivating research into methods for identifying and eliminating them. However, addressing the problem of spurious correlations requires more clarity on what they are and how they arise in language data. Gardner et al (2021) argue that the compositional nature of language implies that \emph{all} correlations between labels and individual "input features" are spurious. This paper analyzes this proposal in the context of a toy example, demonstrating three distinct conditions that can give rise to feature-label correlations in a simple PCFG. Linking the toy example to a structured causal model shows that (1) feature-label correlations can arise even when the label is invariant to interventions on the feature, and (2) feature-label correlations may be absent even when the label is sensitive to interventions on the feature. Because input features will be individually correlated with labels in all but very rare circumstances, domain knowledge must be applied to identify spurious correlations that pose genuine robustness threats.
