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

Neural machine translation (NMT) heavily relies on word-level modelling tolearn semantic representations of input sentences. However, for languageswithout natural word delimiters (e.g., Chinese) where input sentences have tobe tokenized first, conventional NMT is confronted with two issues: 1) it isdifficult to find an optimal tokenization granularity for source sentencemodelling, and 2) errors in 1-best tokenizations may propagate to the encoderof NMT. To handle these issues, we propose word-lattice based Recurrent NeuralNetwork (RNN) encoders for NMT, which generalize the standard RNN to wordlattice topology. The proposed encoders take as input a word lattice thatcompactly encodes multiple tokenizations, and learn to generate new hiddenstates from arbitrarily many inputs and hidden states in preceding time steps.As such, the word-lattice based encoders not only alleviate the negative impactof tokenization errors but also are more expressive and flexible to embed inputsentences. Experiment results on Chinese-English translation demonstrate thesuperiorities of the proposed encoders over the conventional encoder.

Lattice-Based Recurrent Neural Network Encoders for Neural Machine Translation