# Online Versus Offline NMT Quality: An In-depth Analysis on English-German and German-English

arXiv: https://arxiv.org/abs/2006.00814 · HTML: https://mlsanity.com/paper/2006.00814/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 1 Jun 2020.

## Built on by (most cited first, 6 of 6)

- [Over-Generation Cannot Be Rewarded: Length-Adaptive Average Lagging for Simultaneous Speech Translation](https://mlsanity.com/paper/2206.05807/) (2022)
- [Gaussian Multi-head Attention for Simultaneous Machine Translation](https://mlsanity.com/paper/2203.09072/) (2022)
- [Reducing Position Bias in Simultaneous Machine Translation with Length-Aware Framework](https://mlsanity.com/paper/2203.09053/) (2022)
- [Continuous Rating as Reliable Human Evaluation of Simultaneous Speech Translation](https://mlsanity.com/paper/2203.02458/) (2022)
- [Impact of Encoding and Segmentation Strategies on End-to-End Simultaneous Speech Translation](https://mlsanity.com/paper/2104.14470/) (2021)
- [Visualization: the missing factor in Simultaneous Speech Translation](https://mlsanity.com/paper/2111.00514/) (2021)

## 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.