We consider the problem of learning textual entailment models with limited
supervision (5K-10K training examples), and present two complementary
approaches for it. First, we propose knowledge-guided adversarial example
generators for incorporating large lexical resources in entailment models via
only a handful of rule templates. Second, to make the entailment model - a
discriminator - more robust, we propose the first GAN-style approach for
training it using a natural language example generator that iteratively adjusts
based on the discriminator's performance. We demonstrate effectiveness using
two entailment datasets, where the proposed methods increase accuracy by 4.7%
on SciTail and by 2.8% on a 1% training sub-sample of SNLI. Notably, even a
single hand-written rule, negate, improves the accuracy on the negation
examples in SNLI by 6.1%.

本文提出了两种互补的方法来解决在受限监督（5K-10K 训练样本）的情况下学习文本蕴涵模型的问题：一是采用知识引导的对抗样本生成器来通过仅采用少量规则模板将大型词汇资源纳入蕴涵模型中；二是提出了首个采用自然语言例句生成器进行训练的 GAN-style 方法，以使蕴涵模型 - 判别器 - 更加稳健。在两个蕴涵数据集上进行的实证表明，所提出的方法在 SciTail 上将准确度提高了 4.7% ，在 SNLI 的 1% 训练子样本上将准确率提高了 2.8%。值得注意的是，即使是一个手写规则 “否定”，也将 SNLI 中否定样例的准确度提高了 6.1%。