Summariser is a transformer-based tool to generate concise and informative summaries of text data.
Summariser adopts one of multiple text summarisation models to summarise the text that has been retrieved by Retriever.
Implement extractive text summarisation to generate text that summarises research
Select various transformer-based text summarisation models: BART, PEGASUS, T5
Customise by adjusting its parameters and settings, including the length of the summary, the level of detail, and the type of content that is emphasised.
Paraphraser is designed to enable researchers to generate new versions of existing text data that are semantically equivalent but structurally different.
Paraphraser is based on a back-translation strategy, which involves translating the text from the original language to another language and then translating it back again.
Avoiding plagiarism by generating new versions that are syntactically and structurally different.
Identify and preserve the underlying meaning and intent of the original text
Create multiple versions of a document for different audiences
Control the diversity by selecting different languages to be back-translated: French, German, Arabic, Chinese, Russian, Japanese