A practical and feature-rich paraphrasing framework to augment human intents in text form to build robust NLU models for conversational engines. Created by Prithiviraj Damodaran. Open to pull requests and other forms of collaboration.
Paraphrase output for German text is essentially not usable. Is there anything more to be considered? From quickly looking into the paper it looks like it should be possible, but I honestly didn't read deeply into it and I'm no NLP expert.
Paraphrase output for German text is essentially not usable. Is there anything more to be considered? From quickly looking into the paper it looks like it should be possible, but I honestly didn't read deeply into it and I'm no NLP expert.