Work

Attention Strategies for Multi-Source Sequence-to-Sequence Learning

Year: 2017

Type: conference-paper

Abstract: Modeling attention in neural multi-source sequence-to-sequence learning remains a relatively unexplored area, despite its usefulness in tasks that incorporate multiple source languages or modalities.We propose two novel approaches to combine the outputs of attention mechanisms over each source sequence, flat and hierarchical.We compare the proposed methods with existing techniques and present results of systematic evaluation of those methods on the WMT16 Multimodal Translation and Automatic Post-editing tasks.We show that the proposed methods achieve competitive results on both tasks.

Source: Annual Meeting of the Association for Computational Linguistics (ACL)

DOI: 10.18653/v1/p17-2031

Authors: Jindřich Libovický, Jindřich Helcl

Institution: Charles University

Language: English

Topic: Natural Language Processing Techniques

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

Funders: Univerzita Karlova v Praze, Grantová Agentura České Republiky, Horizon 2020 Framework Programme

Awards: G1176565477, G2153309609, Center for Large Scale Multi-modal Data Interpretation