Sequence models are a critical component of modern NLP systems, but their predictions are difficult to explain. We consider model explanations though rationales, subsets of context that can explain individual model predictions. We find sequential rationales by solving a combinatorial optimization: the best rationale is the smallest subset of input tokens that would predict the same output as the full sequence. Enumerating all subsets is intractable, so we propose an efficient greedy algorithm to approximate this objective. The algorithm, which is called greedy rationalization, applies to any model. For this approach to be effective, the model should form compatible conditional distributions when making predictions on incomplete subsets of the context. This condition can be enforced with a short fine-tuning step. We study greedy rationalization on language modeling and machine translation. Compared to existing baselines, greedy rationalization is best at optimizing the combinatorial objective and provides the most faithful rationales. On a new dataset of annotated sequential rationales, greedy rationales are most similar to human rationales.

翻译：序列模型是现代 NLP 系统的关键组成部分, 但是它们的预测很难解释 。 我们考虑模型解释, 其原理, 其背景子集, 可以解释单个模型预测。 我们通过组合优化找到相继原理: 最佳原理是最小的输入符号子集, 可以预测与完整序列相同的输出。 列出所有子集是棘手的, 因此我们建议一种高效的贪婪算法来接近这个目标 。 算法( 称为贪婪合理化) 适用于任何模型 。 要让此方法有效, 模型在对不完整的上下文子集作出预测时, 应该形成兼容的有条件分布 。 这个条件可以通过短小的微调整步骤执行 。 我们研究关于语言建模和机器翻译的贪婪合理化 。 与现有的基线相比, 贪婪合理化是优化组合目标的最佳方法, 并提供最忠实的理由 。 在附加说明的顺序原理的新数据集中, 贪婪的理由与人类的理由非常相似 。