Translate and label!

• Verfasst von: Daza, Angel [VerfasserIn]
• Verfasst von: Frank, Anette [VerfasserIn]
• Titel: Translate and label!
• Titelzusatz: An encoder-decoder approach for cross-lingual semantic role labeling
• Verf.angabe: Angel Daza and Anette Frank
• E-Jahr: 2019
• Jahr: 29 Aug 2019
• Umfang: 13 S.
• Fussnoten: Gesehen am 23.01.2020
• Titel Quelle: Enthalten in: De.arxiv.org
• Ort Quelle: [S.l.] : Arxiv.org, 1991
• Jahr Quelle: 2019
• Band/Heft Quelle: (2019) Artikel-Nummer 1908.11326, 13 Seiten
• Datenträger: Online-Ressource
• Sprache: eng
• Bibliogr. Hinweis: Forschungsdaten: Daza, Angel: Encoder-decoder model for semantic role labeling
• Sach-SW: Computer Science - Computation and Language
• Sach-SW: Computer Science - Machine Learning
• K10plus-PPN: 1688153624

Abstract

We propose a Cross-lingual Encoder-Decoder model that simultaneously translates and generates sentences with Semantic Role Labeling annotations in a resource-poor target language. Unlike annotation projection techniques, our model does not need parallel data during inference time. Our approach can be applied in monolingual, multilingual and cross-lingual settings and is able to produce dependency-based and span-based SRL annotations. We benchmark the labeling performance of our model in different monolingual and multilingual settings using well-known SRL datasets. We then train our model in a cross-lingual setting to generate new SRL labeled data. Finally, we measure the effectiveness of our method by using the generated data to augment the training basis for resource-poor languages and perform manual evaluation to show that it produces high-quality sentences and assigns accurate semantic role annotations. Our proposed architecture offers a flexible method for leveraging SRL data in multiple languages.