Table 7 
BLEU scores and perplexity for eval-AWD-LSTM-based evaluation on texts generated from the COCO image data set by different GAN models: SeqGAN (Yu et al. 2017), MaliGAN (Che et al. 2017), RankGAN (Lin et al. 2017), LeakGAN (Guo et al. 2018), and TextGAN (Zhang et al. 2017).
 SeqGANMaliGANRankGANLeakGANTextGANMLEImageCoco
BLEU-2 0.92 0.89 0.94 0.93 0.65 0.92 1.00 
BLEU-3 0.75 0.70 0.80 0.82 0.65 0.68 1.00 
BLEU-4 0.53 0.48 0.60 0.66 0.60 0.57 1.00 
BLEU-5 0.35 0.31 0.41 0.47 0.52 0.39 1.00 
eval-AWD-LSTM 179.29 272.53 132.90 146.26 129.93 176.34 44.17 
 SeqGANMaliGANRankGANLeakGANTextGANMLEImageCoco
BLEU-2 0.92 0.89 0.94 0.93 0.65 0.92 1.00 
BLEU-3 0.75 0.70 0.80 0.82 0.65 0.68 1.00 
BLEU-4 0.53 0.48 0.60 0.66 0.60 0.57 1.00 
BLEU-5 0.35 0.31 0.41 0.47 0.52 0.39 1.00 
eval-AWD-LSTM 179.29 272.53 132.90 146.26 129.93 176.34 44.17 
View Large
Close Modal
This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal
This site uses cookies. By continuing to use our website, you are agreeing to our privacy policy. No content on this site may be used to train artificial intelligence systems without permission in writing from the MIT Press.
Accept