论文标题
通过图案利用培训来增强表格推理
Enhancing Tabular Reasoning with Pattern Exploiting Training
论文作者
论文摘要
基于预先训练的语言模型的最新方法比表格任务表现出优越的性能（例如，表格NLI），尽管显示出固有的问题，例如不使用正确的证据和对输入的预测不一致，同时通过表格数据进行推理。在这项工作中，我们利用了先前训练的语言模型上的模式开发培训（PET）（即战略传销）来增强这些表格推理模型的先前存在的知识和推理能力。与当前的基线相比，我们升级的模型对知识事实和表格推理具有更高的了解。此外，我们证明了此类模型对于在信息磁盘上进行表格推断的下游任务更有效。此外，我们展示了模型对通过各种字符和单词级扰动产生的对抗集的鲁棒性。
Recent methods based on pre-trained language models have exhibited superior performance over tabular tasks (e.g., tabular NLI), despite showing inherent problems such as not using the right evidence and inconsistent predictions across inputs while reasoning over the tabular data. In this work, we utilize Pattern-Exploiting Training (PET) (i.e., strategic MLM) on pre-trained language models to strengthen these tabular reasoning models' pre-existing knowledge and reasoning abilities. Our upgraded model exhibits a superior understanding of knowledge facts and tabular reasoning compared to current baselines. Additionally, we demonstrate that such models are more effective for underlying downstream tasks of tabular inference on InfoTabs. Furthermore, we show our model's robustness against adversarial sets generated through various character and word level perturbations.
