BriefGPT.xyz
Jun, 2018
iParaphrasing：通过图像提取基于视觉的复述句
iParaphrasing: Extracting Visually Grounded Paraphrases via an Image
HTML
PDF
Chenhui Chu, Mayu Otani, Yuta Nakashima
TL;DR
本文提出了一种新的任务iParaphrasing，通过提取基于视觉的复述词（VGPs）来改善语言和图像多模态任务的性能，使用各种现有方法和基于神经网络的图像注意力方法建模VGPs之间的相似性并报告了结果。
Abstract
A
paraphrase
is a restatement of the meaning of a text in other words. Paraphrases have been studied to enhance the performance of many
natural language processing
tasks. In this paper, we propose a novel task iP
→