SeqGAN: Sequence Generative Adversarial Nets with Policy Gradient

18 Sep 2016Lantao YuWeinan ZhangJun WangYong Yu

As a new way of training generative models, Generative Adversarial Nets (GAN) that uses a discriminative model to guide the training of the generative model has enjoyed considerable success in generating real-valued data. However, it has limitations when the goal is for generating sequences of discrete tokens... (read more)

PDF Abstract

Evaluation results from the paper


Task Dataset Model Metric name Metric value Global rank Compare
Text Generation Chinese Poems SeqGAN BLEU-2 0.738 # 3
Text Generation COCO Captions SeqGAN BLEU-2 0.831 # 4
Text Generation COCO Captions SeqGAN BLEU-3 0.642 # 4
Text Generation COCO Captions SeqGAN BLEU-4 0.521 # 3
Text Generation COCO Captions SeqGAN BLEU-5 0.427 # 3
Text Generation EMNLP2017 WMT SeqGAN BLEU-2 0.859 # 2
Text Generation EMNLP2017 WMT SeqGAN BLEU-3 0.6015 # 2
Text Generation EMNLP2017 WMT SeqGAN BLEU-4 0.4541 # 2
Text Generation EMNLP2017 WMT SeqGAN BLEU-5 0.4498 # 3