Using Sentence Plausibility to Learn the Semantics of Transitive Verbs
Tamara Polajnar Laura Rimell Stephen Clark Computer Laboratory University of Cambridge United Kingdom
Abstract
The functional approach to compositional distributional semantics considers transitive verbs to be linear maps that transform the distributional vectors representing nouns into a vector representing a sentence. We conduct an initial investigation that uses a matrix consisting of the parameters of a logistic regression classifier trained on a plausibility task as a transitive verb function. We compare our method to a commonly used corpus-based method for constructing a verb matrix and find that the plausibility training may be more effective for disambiguation tasks.
中文速览
如何让计算机理解"animals eat plants"和"animals eat planets"之间的语义差异，是组合分布语义学（compositional distributional semantics）的核心挑战之一。研究者将及物动词表示为一个矩阵，分别尝试了三种学习方式：一种是传统的基于语料库共现统计的分布式方法，另一种是用逻辑回归分类器在"合理/不合理句子"区分任务上训练得到的矩阵，以及只用正例训练的折中方案。实验在动词消歧和句子相似度两个任务上对比发现，基于合理性训练的回归方法在消歧任务上更有优势，而传统分布式方法在句子相似度任务上表现更好，原因在于两类任务对语义空间的需求本质上不同——消歧依赖合理性判断，相似度则更依赖话题信息。这一发现提示我们，针对不同下游任务需要构建不同性质的句子语义空间，为后续低成本、任务适配的句子表示学习指明了方向。
原文 arXiv:1411.7942;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1411.7942v2