arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1902.10296cs.CL
收藏

从文本中解码事件相关电位的框架

A Framework for Decoding Event-Related Potentials from Text

University of Rochester(罗切斯特大学);Department of Brain and Cognitive Sciences(脑与认知科学系)

Shaorong Yan, Aaron Steven White

摘要:

我们提出了一个新的框架,用于对阅读过程中收集的事件相关电位(ERP)进行建模,该框架将预训练的卷积解码器与语言模型相结合。使用这个框架,我们比较各种现有的和新的句子处理模型的能力,以重建ERPs。我们发现,现代语境词嵌入表现不佳的基于语料库的模型,但结合起来,这两个优于任何一个本身。

英文摘要:

We propose a novel framework for modeling event-related potentials (ERPs) collected during reading that couples pre-trained convolutional decoders with a language model. Using this framework, we compare the abilities of a variety of existing and novel sentence processing models to reconstruct ERPs. We find that modern contextual word embeddings underperform surprisal-based models but that, combined, the two outperform either on its own.

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。

↑