arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1909.04849cs.CLcs.AI
收藏

弱监督问答的离散硬EM方法

A Discrete Hard EM Approach for Weakly Supervised Question Answering

University of Washington(华盛顿大学);Princeton University(普林斯顿大学);Facebook AI Research(Facebook人工智能研究院);Allen Institute for Artificial Intelligence(艾伦人工智能研究所)

Sewon Min, Danqi Chen, Hannaneh Hajishirzi, Luke Zettlemoyer

摘要:

许多问答(QA)任务只提供弱监督的答案应该如何计算。例如,TriviaQA答案是可以在支持文档中多次提及的实体,而DROP答案可以通过从参考文本中的数字导出许多不同的方程来计算。在本文中,我们证明了将这些任务转换为离散潜变量学习问题是可能的,这些问题具有预先计算的特定于任务的可能“解决方案”(例如,不同的提及或方程),其中包含一个正确的选项。然后,我们开发了一个硬EM学习方案,在每次更新时计算相对于最可能的解决方案的梯度。尽管它的简单性,我们表明,这种方法显着优于以前的方法在六个QA任务,包括2- 10%的绝对增益,并实现了国家的最先进的其中五个。使用硬更新而不是最大化边际可能性是这些结果的关键,因为它鼓励模型找到一个正确的答案,我们通过详细的定性分析显示。

英文摘要:

Many question answering (QA) tasks only provide weak supervision for how the answer should be computed. For example, TriviaQA answers are entities that can be mentioned multiple times in supporting documents, while DROP answers can be computed by deriving many different equations from numbers in the reference text. In this paper, we show it is possible to convert such tasks into discrete latent variable learning problems with a precomputed, task-specific set of possible "solutions" (e.g. different mentions or equations) that contains one correct option. We then develop a hard EM learning scheme that computes gradients relative to the most likely solution at each update. Despite its simplicity, we show that this approach significantly outperforms previous methods on six QA tasks, including absolute gains of 2--10%, and achieves the state-of-the-art on five of them. Using hard updates instead of maximizing marginal likelihood is key to these results as it encourages the model to find the one correct answer, which we show through detailed qualitative analysis.

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。

↑