Few-shot learning (FSL) aims to learn new categories with a few visual samples per class. Few-shot class representations are often biased due to data scarcity. To mitigate this issue, the authors propose to generate visual samples based on semantic embeddings using a conditional variational autoencoder (CVAE) model. This CVAE model is trained on base classes and used to generate features for novel classes. More importantly, the authors guide this VAE to strictly generate representative samples by removing non-representative samples from the base training set when training the CVAE model. The work shows that this training scheme enhances the representativeness of the generated samples and therefore, improves the few-shot classification results. Experimental results show that the proposed method improves three FSL baseline methods by substantial margins, achieving state-of-the-art few-shot classification performance on miniImageNet and tieredImageNet datasets for both 1-shot and 5-shot settings.