FlowSeq: Non-Autoregressive Conditional Sequence Generation with Generative Flow

Xuezhe MaChunting ZhouXian LiGraham NeubigEduard Hovy

   Papers with code   Abstract  PDF

Most sequence-to-sequence (seq2seq) models are autoregressive; they generate each token by conditioning on previously generated tokens. In contrast, non-autoregressive seq2seq models generate all tokens in one pass, which leads to increased efficiency through parallel processing on hardware such as GPUs... (read more)

Benchmarked Models

No benchmarked models yet. Click here to submit a model.