arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2005.12116cs.CL
收藏

NILE:具有忠实自然语言解释的自然语言推理

NILE : Natural Language Inference with Faithful Natural Language Explanations

Indian Institute of Science(印度科学学院)

Sawan Kumar, Partha Talukdar

摘要:

最近,深度学习模型在NLP分类任务中的流行和成功伴随着对预测标签生成某种形式的自然语言解释的需求。期望这样生成的自然语言(NL)解释是忠实的,即,它们应该与模型的内部决策很好地关联。在这项工作中,我们专注于自然语言推理(NLI)的任务,并解决以下问题:我们可以建立NLI系统,产生高精度的标签,同时也产生忠实的解释其决定?我们提出了自然语言推理的标签特定的解释(NILE),一种新的NLI方法,利用自动生成的标签特定的NL解释,以产生标签及其忠实的解释。我们证明了NILE的有效性,通过自动化和人工评价所产生的标签和解释比以前报道的方法。我们对NILE的评估也支持这样的说法,即可以设计出能够为他们的决定提供可测试解释的精确系统。我们讨论了NILE的解释的忠实性方面的敏感性的决定,相应的解释。我们认为,除了标签和解释准确性外,对忠实性的明确评估是评估模型解释的重要步骤。此外,我们表明,任务特定的探针是必要的,以建立这样的灵敏度。

英文摘要:

The recent growth in the popularity and success of deep learning models on NLP classification tasks has accompanied the need for generating some form of natural language explanation of the predicted labels. Such generated natural language (NL) explanations are expected to be faithful, i.e., they should correlate well with the model's internal decision making. In this work, we focus on the task of natural language inference (NLI) and address the following question: can we build NLI systems which produce labels with high accuracy, while also generating faithful explanations of its decisions? We propose Natural-language Inference over Label-specific Explanations (NILE), a novel NLI method which utilizes auto-generated label-specific NL explanations to produce labels along with its faithful explanation. We demonstrate NILE's effectiveness over previously reported methods through automated and human evaluation of the produced labels and explanations. Our evaluation of NILE also supports the claim that accurate systems capable of providing testable explanations of their decisions can be designed. We discuss the faithfulness of NILE's explanations in terms of sensitivity of the decisions to the corresponding explanations. We argue that explicit evaluation of faithfulness, in addition to label and explanation accuracy, is an important step in evaluating model's explanations. Further, we demonstrate that task-specific probes are necessary to establish such sensitivity.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑