A novel repetition normalized adversarial reward for headline generation
This work addresses the issue of low-quality text generation in headline generation for natural language processing applications, representing an incremental improvement over existing methods.
The paper tackled the problem of incoherent and repetitive phrase generation in reinforcement learning for language models by proposing a repetition normalized adversarial reward, resulting in significant improvements including a 3.24 increase in ROUGE-1, a 2.25 increase in ROUGE-L, and a 4.98% decrease in repetition rate.
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\%).