In order to reveal the rationale behind model predictions, many works have exploited providing explanations in various forms. Recently, to further guarantee readability, more and more works turn to generate sentence-level human language explanations. However, current works pursuing sentence-level explanations rely heavily on annotated training data, which limits the development of interpretability to only a few tasks. As far as we know, this paper is the first to explore this problem smoothly from weak-supervised learning to unsupervised learning. Besides, we also notice the high latency of autoregressive sentence-level explanation generation, which leads to asynchronous interpretability after prediction. Therefore, we propose a non-autoregressive interpretable model to facilitate parallel explanation generation and simultaneous prediction. Through extensive experiments on Natural Language Inference task and Spouse Prediction task, we find that users are able to train classifiers with comparable performance $10-15\times$ faster with parallel explanation generation using only a few or no annotated training data.
翻译:为了揭示模型预测背后的理由,许多作品都利用了多种形式的解释。最近,为了进一步保证可读性,越来越多的作品转而产生判决一级的人类语言解释。然而,目前追求判决一级的解释在很大程度上依赖于附加说明的培训数据,这限制了解释性发展,只局限于少数任务。据我们所知,本文件是第一个顺利探讨这一问题的人,从监督不力的学习到不受监督的学习。此外,我们还注意到自动递减性判决一级解释生成的高度长期性,这导致预测后的非同步性解释性。因此,我们提出一个非递性解释性的解释模型,以便利平行的解释生成和同步预测。通过对自然语言推论任务和配偶预测任务的广泛实验,我们发现用户能够用少量或没有附加说明的培训数据,对类似性能10-15小时的分类人员进行培训,并更快地同时进行解释生成数据。