Neural Headline Generation with Sentence-wise Optimization
arXiv:1604.01904
Abstract
Recently, neural models have been proposed for headline generation by learning to map documents to headlines with recurrent neural networks. Nevertheless, as traditional neural network utilizes maximum likelihood estimation for parameter optimization, it essentially constrains the expected training objective within word level rather than sentence level. Moreover, the performance of model prediction significantly relies on training data distribution. To overcome these drawbacks, we employ minimum risk training strategy in this paper, which directly optimizes model parameters in sentence level with respect to evaluation metrics and leads to significant improvements for headline generation. Experiment results show that our models outperforms state-of-the-art systems on both English and Chinese headline generation tasks.
References in corpus (5)
Cited by in corpus (13)
- Convolutional Sequence to Sequence Learning
- Minimum Risk Training for Neural Machine Translation
- A Reinforced Topic-Aware Convolutional Sequence-to-Sequence Model for Abstractive Text Summarization
- Towards Explainable NLP: A Generative Explanation Framework for Text Classification
- On the Weaknesses of Reinforcement Learning for Neural Machine Translation
- Hooks in the Headline: Learning to Generate Headlines with Controlled Styles
- Language Modeling with Sparse Product of Sememe Experts
- Diverse, Controllable, and Keyphrase-Aware: A Corpus and Method for News Multi-Headline Generation
- Contrastive Attention Mechanism for Abstractive Sentence Summarization
- Self-organized Hierarchical Softmax
- A Hybrid Word-Character Approach to Abstractive Summarization
- GRET: Global Representation Enhanced Transformer
- Sentence-wise Smooth Regularization for Sequence to Sequence Learning