Paraphrase Detection on Noisy Subtitles in Six Languages
This work addresses paraphrase detection for multilingual subtitle processing, but it is incremental as it applies existing methods to a new noisy dataset.
The study tackled paraphrase detection on noisy subtitle data in six European languages, finding that a gated recurrent averaging network (GRAN) model outperformed a word-averaging model and was more robust to noise, with better results from more and noisier data.
We perform automatic paraphrase detection on subtitle data from the Opusparcus corpus comprising six European languages: German, English, Finnish, French, Russian, and Swedish. We train two types of supervised sentence embedding models: a word-averaging (WA) model and a gated recurrent averaging network (GRAN) model. We find out that GRAN outperforms WA and is more robust to noisy training data. Better results are obtained with more and noisier data than less and cleaner data. Additionally, we experiment on other datasets, without reaching the same level of performance, because of domain mismatch between training and test data.