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

Due to Multilingual Neural Machine Translation's (MNMT) capability ofzero-shot translation, many works have been carried out to fully exploit thepotential of MNMT in zero-shot translation. It is often hypothesized thatpositional information may hinder the MNMT from outputting a robust encodedrepresentation for decoding. However, previous approaches treat all thepositional information equally and thus are unable to selectively removecertain positional information. In sharp contrast, this paper investigates howto learn to selectively preserve useful positional information.  We describe the specific mechanism of positional information influencing MNMTfrom the perspective of linguistics at the token level. We design a token-levelposition disentangle module (TPDM) framework to disentangle positionalinformation at the token level based on the explanation. Our experimentsdemonstrate that our framework improves zero-shot translation by a large marginwhile reducing the performance loss in the supervised direction compared toprevious works.

TPDM: Selectively Removing Positional Information for Zero-shot Translation via Token-Level Position Disentangle Module