文章基本信息

标题：Does the Order of Training Samples Matter? Improving Neural Data-to-Text Generation with Curriculum Learning
本地全文：下载
作者：Ernie Chang ; Hui-Syuan Yeh ; Vera Demberg 等
期刊名称：Conference on European Chapter of the Association for Computational Linguistics (EACL)
出版年度：2021
卷号：2021
页码：727-733
DOI：10.18653/v1/2021.eacl-main.61
语种：English
出版社：ACL Anthology
摘要：Recent advancements in data-to-text generation largely take on the form of neural end-to-end systems. Efforts have been dedicated to improving text generation systems by changing the order of training samples in a process known as curriculum learning. Past research on sequence-to-sequence learning showed that curriculum learning helps to improve both the performance and convergence speed. In this work, we delve into the same idea surrounding the training samples consisting of structured data and text pairs, where at each update, the curriculum framework selects training samples based on the model’s competence. Specifically, we experiment with various difficulty metrics and put forward a soft edit distance metric for ranking training samples. On our benchmarks, it shows faster convergence speed where training time is reduced by 38.7% and performance is boosted by 4.84 BLEU.