A framework for evaluating and explaining the robustness of NLP models

在NLP任务中评估监督机器学习模型泛化的标准做法是使用以前看不见的（即保留的）数据，并使用各种指标（如准确性）报告其性能。虽然举行了数据的度量报告总结了模型的性能，最终这些结果代表基准的汇总统计数据，并不反映在模型的行为和鲁棒性的细微差别，当应用在现实世界的systems.We提出了一个鲁棒性评估框架NLP模型关注的参数和事实，其中包括解释鲁棒性失败，以支持系统和有效的评估。我们将开发新的方法来模拟来自现有数据集的真实文本，以帮助评估模型在野外部署时的稳定性和一致性。模拟方法将用于通过基于文本的转换和数据集以及捕获语言模式的数据子集的分布变化来挑战NLP模型，以提供对真实世界语言现象的系统覆盖。此外，我们的框架将摆脱洞察模型的鲁棒性，通过生成鲁棒性故障的解释沿着词汇，形态和语法维度，从各种数据集模拟和数据子集中提取，从而脱离目前的方法，仅提供一个度量来量化鲁棒性。我们将专注于两个NLP研究领域，论点挖掘和事实验证，然而，一些模拟方法和鲁棒性解释也可扩展到其他NLP任务。

The standard practice for evaluating the generalisation of supervised machine learning models in NLP tasks is to use previously unseen (i.e. held-out) data and report the performance on it using various metrics such as accuracy. Whilst metrics reported on held-out data summarise a model's performance, ultimately these results represent aggregate statistics on benchmarks and do not reflect the nuances in model behaviour and robustness when applied in real-world systems.We propose a robustness evaluation framework for NLP models concerned with arguments and facts, which encompasses explanations for robustness failures to support systematic and efficient evaluation. We will develop novel methods for simulating real-world texts stemming from existing datasets, to help evaluate the stability and consistency of models when deployed in the wild. The simulation methods will be used to challenge NLP models through text-based transformations and distribution shifts on datasets as well as on data sub-sets that capture linguistic patterns, to provide a systematic coverage of real-world linguistic phenomena. Furthermore, our framework will shed insights into a model's robustness by generating explanations for robustness failures along the lexical, morphological, and syntactic dimensions, extracted from the various dataset simulations and data sub-sets, thus departing from current approaches that solely provide a metric to quantify robustness. We will focus on two NLP research areas, argument mining and fact verification, however, several simulation methods and the robustness explanations are also scalable to other NLP tasks.