papers AI Learner
The Github is limit! Click to go to the new site.

A novel repetition normalized adversarial reward for headline generation

2019-02-19
Peng Xu, Pascale Fung

Abstract

While reinforcement learning can effectively improve language generation models, it often suffers from generating incoherent and repetitive phrases \cite{paulus2017deep}. In this paper, we propose a novel repetition normalized adversarial reward to mitigate these problems. Our repetition penalized reward can greatly reduce the repetition rate and adversarial training mitigates generating incoherent phrases. Our model significantly outperforms the baseline model on ROUGE-1\,(+3.24), ROUGE-L\,(+2.25), and a decreased repetition-rate (-4.98\%).

Abstract (translated by Google)
URL

http://arxiv.org/abs/1902.07110

PDF

http://arxiv.org/pdf/1902.07110


Similar Posts

Comments