FlowSeq: Non-Autoregressive Conditional Sequence Generation with Generative Flow
This addresses the efficiency problem in sequence-to-sequence tasks like machine translation by enabling parallel decoding, though it is incremental as it builds on existing non-autoregressive methods.
The paper tackles the challenge of non-autoregressive sequence generation, which lags in accuracy compared to autoregressive models, by proposing FlowSeq, a model using generative flow for latent variable modeling, achieving comparable performance to state-of-the-art non-autoregressive models on three NMT benchmarks with almost constant decoding time relative to sequence length.
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. However, directly modeling the joint distribution of all tokens simultaneously is challenging, and even with increasingly complex model structures accuracy lags significantly behind autoregressive models. In this paper, we propose a simple, efficient, and effective model for non-autoregressive sequence generation using latent variable models. Specifically, we turn to generative flow, an elegant technique to model complex distributions using neural networks, and design several layers of flow tailored for modeling the conditional density of sequential latent variables. We evaluate this model on three neural machine translation (NMT) benchmark datasets, achieving comparable performance with state-of-the-art non-autoregressive NMT models and almost constant decoding time w.r.t the sequence length.