Machine Translation Aided Bilingual Data-to-Text Generation and Semantic Parsing
2020-12-01ACL (WebNLG, INLG) 2020Unverified0· sign in to hype
Oshin Agarwal, Mihir Kale, Heming Ge, Siamak Shakeri, Rami Al-Rfou
Unverified — Be the first to reproduce this paper.
ReproduceAbstract
We present a system for bilingual Data-ToText Generation and Semantic Parsing. We use a text-to-text generator to learn a single model that works for both languages on each of the tasks. The model is aided by machine translation during both pre-training and fine-tuning. We evaluate the system on WebNLG 2020 data 1 , which consists of RDF triples in English and natural language sentences in English and Russian for both the tasks. We achieve considerable gains over monolingual models, especially on unseen relations and Russian.