Input Combination Strategies for Multi-Source Transformer Decoder
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F18%3A10390178" target="_blank" >RIV/00216208:11320/18:10390178 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
Input Combination Strategies for Multi-Source Transformer Decoder
Original language description
In multi-source sequence-to-sequence tasks, the attention mechanism can be modeled in several ways. This topic has been thoroughly studied on recurrent architectures. In this paper, we extend the previous work to the encoder-decoder attention in the Transformer architecture. We propose four different input combination strategies for the encoder-decoder attention: serial, parallel, flat, and hierarchical. We evaluate our methods on tasks of multimodal translation and translation with multiple source languages. The experiments show that the models are able to use multiple sources and improve over single source baselines.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
—
OECD FORD branch
10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
Result continuities
Project
Result was created during the realization of more than one project. More information in the Projects tab.
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)
Others
Publication year
2018
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
Proceedings of the Third Conference on Machine Translation, Volume 1: Research Papers
ISBN
978-1-948087-81-0
ISSN
—
e-ISSN
neuvedeno
Number of pages
8
Pages from-to
253-260
Publisher name
Association for Computational Linguistics
Place of publication
Stroudsburg, PA, USA
Event location
Bruxelles, Belgium
Event date
Oct 31, 2018
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—