Counterfactuals to Control Latent Disentangled Text Representations for Style Transfer
| Autor: | Nikhil Kaushik, Niyati Chhaya, Harshit Nyati, Sopan Khosla, Sharmila Reddy Nangi |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Counterfactual thinking
Counterfactual conditional business.industry Computer science Formality computer.software_genre Style (sociolinguistics) Simple (abstract algebra) Artificial intelligence Control (linguistics) Representation (mathematics) business computer Natural language processing Sentence |
| Zdroj: | ACL/IJCNLP (2) |
| DOI: | 10.18653/v1/2021.acl-short.7 |
| Popis: | Disentanglement of latent representations into content and style spaces has been a commonly employed method for unsupervised text style transfer. These techniques aim to learn the disentangled representations and tweak them to modify the style of a sentence. In this paper, we propose a counterfactual-based method to modify the latent representation, by posing a ‘what-if’ scenario. This simple and disciplined approach also enables a fine-grained control on the transfer strength. We conduct experiments with the proposed methodology on multiple attribute transfer tasks like Sentiment, Formality and Excitement to support our hypothesis. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|