Differentiable N-gram Objective on Abstractive Summarization

1 code implementation • 8 Feb 2022 • Yunqi Zhu, Xuebing Yang, Yuanyuan Wu, Mingjin Zhu, Wensheng Zhang

ROUGE is a standard automatic evaluation metric based on n-grams for sequence-to-sequence tasks, while cross-entropy loss is an essential objective of neural network language model that optimizes at a unigram level.

Abstractive Text Summarization Language Modelling