Bottleneck (context vector bottleneck)

Appears in 1 paper

The design flaw in plain seq2seq (Paper 06): the entire source sentence's meaning must be compressed into a single fixed-size vector before decoding can begin.

As used in Paper 07 — Neural Machine Translation by Jointly Learning to Align and Translate →

The design flaw in plain seq2seq (Paper 06): the entire source sentence's meaning must be compressed into a single fixed-size vector before decoding can begin. For long sentences, important information is lost during this compression.