Concept
Seq2seq
seq2seq (sequence-to-sequence) is a model architecture that maps an input sequence of tokens to an output sequence of tokens, where the two sequences can differ in length and vocabulary. The canonical shape is an encoder that reads the full input into a hidden…
The rest of “Seq2seq” is a premium feature: every concept in the library gets a precise, practitioner-focused write-up like this one, cross-linked straight from the paper summaries that use it.
Log in to unlock→