Online Versus Offline NMT Quality: An In-depth Analysis on English-German and German-English
Not on the map. Why?
Abstract (arXiv)
We conduct in this work an evaluation study comparing offline and online neural machine translation architectures. Two sequence-to-sequence models: convolutional Pervasive Attention (Elbayad et al. 2018) and attention-based Transformer (Vaswani et al. 2017) are considered. We investigate, for both architectures, the impact of online decoding constraints on the translation quality through a carefully designed human evaluation on English-German and German-English language pairs, the latter being particularly sensitive to latency constraints. The evaluation results allow us to identify the strengths and shortcomings of each model when we shift to the online setup.
Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). First posted on arXiv 1 Jun 2020.
Builds on
No reference list yet.
Built on by
Citing papers per year: 2021: 2; 2022: 4.
6 papers.
- 2022Over-Generation Cannot Be Rewarded: Length-Adaptive Average Lagging for Simultaneous Speech Translationcited by 56
- 2022Gaussian Multi-head Attention for Simultaneous Machine Translationcited by 17
- 2022Reducing Position Bias in Simultaneous Machine Translation with Length-Aware Frameworkcited by 16
- 2022Continuous Rating as Reliable Human Evaluation of Simultaneous Speech Translationcited by 6
- 2021Impact of Encoding and Segmentation Strategies on End-to-End Simultaneous Speech Translationcited by 4
- 2021Visualization: the missing factor in Simultaneous Speech Translationcited by 2
- 2022Over-Generation Cannot Be Rewarded: Length-Adaptive Average Lagging for Simultaneous Speech Translation12 Jun 2022
- 2022Gaussian Multi-head Attention for Simultaneous Machine Translation17 Mar 2022
- 2022Reducing Position Bias in Simultaneous Machine Translation with Length-Aware Framework17 Mar 2022
- 2022Continuous Rating as Reliable Human Evaluation of Simultaneous Speech Translation4 Mar 2022
- 2021Visualization: the missing factor in Simultaneous Speech Translation31 Oct 2021
- 2021Impact of Encoding and Segmentation Strategies on End-to-End Simultaneous Speech Translation29 Apr 2021