In neural machine translation, the decoder handles an easier but more noise-sensitive task than the encoder, because it depends strongly on the immediately preceding output words.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Hard but Robust, Easy but Sensitive: How Encoder and Decoder Perform in Neural Machine Translation
In neural machine translation, the decoder handles an easier but more noise-sensitive task than the encoder, because it depends strongly on the immediately preceding output words.