Don't put volumes

<br> NLG Task<br/> <br>Cross-Lingual Summarization from english to chinese<br/> <br>Training Data<br/> <br>– 62256.2 Dataset Gigaword DUC2004 D LCSTS CNN/DM<br/> <br>Model Description<br/> <br>transformer architectures is used with ScaleNorm normalisation. The mapper is a linear layer with a hidden size of 512, and the discriminator is a two-layer linear layer with a hidden size of 2048. Several loss functions defined <br/> <br>Key Contribution<br/> <br>First unsupervised cross lingual summarisation<br/>