SBSG model generates sequences bidirectionally from ends to middle via interactive attention, claiming faster decoding and better quality than autoregressive Transformer on NMT and summarization tasks.
Convolutional sequence to sequence learning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Sequence Generation: From Both Sides to the Middle
SBSG model generates sequences bidirectionally from ends to middle via interactive attention, claiming faster decoding and better quality than autoregressive Transformer on NMT and summarization tasks.