Corpus ID: 167217319

Unsupervised Controllable Text Generation with Global Variation Discovery and Disentanglement

@article{Xu2019UnsupervisedCT,
  title={Unsupervised Controllable Text Generation with Global Variation Discovery and Disentanglement},
  author={Peng Xu and Yanshuai Cao and J. C. K. Cheung},
  journal={ArXiv},
  year={2019},
  volume={abs/1905.11975}
}
Existing controllable text generation systems rely on annotated attributes, which greatly limits their capabilities and applications. [...] Key Method We do so by decomposing the latent space of the VAE into two parts: one incorporates structural constraints to capture dominant global variations implicitly present in the data, e.g., sentiment or topic; the other is unstructured and is used for the reconstruction of the source sentences.Expand
Stylized Text Generation: Approaches and Applications
Formality Style Transfer with Shared Latent Space
Enhancing Controllability of Text Generation
Style Example-Guided Text Generation using Generative Adversarial Transformers
Do Sequence-to-sequence VAEs Learn Global Features of Sentences?
Exploring Controllable Text Generation Techniques

References

SHOWING 1-10 OF 20 REFERENCES
Toward Controlled Generation of Text
Unsupervised Text Style Transfer using Language Models as Discriminators
Language Models are Unsupervised Multitask Learners
beta-VAE: Learning Basic Visual Concepts with a Constrained Variational Framework
Generating Sentences from a Continuous Space
Topic Compositional Neural Language Model
Adversarially Regularized Autoencoders
Deep contextualized word representations
...
1
2
...