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 2103.07875cs.CL
收藏

基于句级噪声对比估计的词级语言模型学习

Learning a Word-Level Language Model with Sentence-Level Noise Contrastive Estimation for Contextual Sentence Probability Estimation

Seoul National University(首尔大学)

Heewoong Park, Sukhyun Cho, Jonghun Park

摘要:

推断句子或词序列的概率分布是自然语言处理中的一个关键过程。虽然单词级语言模型(LM)已被广泛采用的计算单词序列的联合概率,他们有困难,在捕捉足够长的上下文句子概率估计(SPE)。为了克服这一点,最近的研究引入了使用递归神经网络(RNN)的重复级噪声对比估计(NCE)的训练方法。在这项工作中,我们试图扩展它的上下文SPE,其目的是估计一个条件句的概率给定以前的文本。所提出的NCE独立于先前文本对否定句进行采样,使得训练后的模型对与\textcolor{blue}{the}上下文更一致的句子给出更高的概率。我们将我们的方法应用于一个简单的单词级RNN LM,以关注单词级NCE训练的效果,而不是网络架构。评估的质量对多项选择完形填空式的问题,包括人类和自动生成的问题。实验结果表明,该方法提高了词级RNN LM的SPE质量。

英文摘要:

Inferring the probability distribution of sentences or word sequences is a key process in natural language processing. While word-level language models (LMs) have been widely adopted for computing the joint probabilities of word sequences, they have difficulty in capturing a context long enough for sentence probability estimation (SPE). To overcome this, recent studies introduced training methods using sentence-level noise-contrastive estimation (NCE) with recurrent neural networks (RNNs). In this work, we attempt to extend it for contextual SPE, which aims to estimate a conditional sentence probability given a previous text. The proposed NCE samples negative sentences independently of a previous text so that the trained model gives higher probabilities to the sentences that are more consistent with \textcolor{blue}{the} context. We apply our method to a simple word-level RNN LM to focus on the effect of the sentence-level NCE training rather than on the network architecture. The quality of estimation was evaluated against multiple-choice cloze-style questions including both human and automatically generated questions. The experimental results show that the proposed method improved the SPE quality for the word-level RNN LM.

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。

↑