Retrieval-Augmented Multilingual Keyphrase Generation with Retriever-Generator Iterative Training. (arXiv:2205.10471v2 [cs.CL] UPDATED)

Keyphrase generation is the task of automatically predicting keyphrases given
a piece of long text. Despite its recent flourishing, keyphrase generation on
non-English languages haven't been vastly investigated. In this paper, we call
attention to a new setting named multilingual keyphrase generation and we
contribute two new datasets, EcommerceMKP and AcademicMKP, covering six
languages. Technically, we propose a retrieval-augmented method for
multilingual keyphrase generation to mitigate the data shortage problem in
non-English languages. The retrieval-augmented model leverages keyphrase
annotations in …

arxiv generation generator iterative retrieval training