English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Recent work has shown that deeper character-based neural machine translation(NMT) models can outperform subword-based models. However, it is still unclearwhat makes deeper character-based models successful. In this paper, we conductan investigation into pure character-based models in the case of translatingFinnish into English, including exploring the ability to learn word senses andmorphological inflections and the attention mechanism. We demonstrate thatword-level information is distributed over the entire character sequence ratherthan over a single character, and characters at different positions playdifferent roles in learning linguistic knowledge. In addition, character-basedmodels need more layers to encode word senses which explains why only deepermodels outperform subword-based models. The attention distribution patternshows that separators attract a lot of attention and we explore a sparseword-level attention to enforce character hidden states to capture the fullword-level information. Experimental results show that the word-level attentionwith a single head results in 1.2 BLEU points drop.

Understanding Pure Character-Based Neural Machine Translation: The Case of Translating Finnish into English