# Exploiting Linguistic Resources for Neural Machine Translation Using Multi-task Learning

arXiv: https://arxiv.org/abs/1708.00993 · HTML: https://mlsanity.com/paper/1708.00993/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Aug 2017.

## Built on by (most cited first, 11 of 11)

- [Survey of Low-Resource Machine Translation](https://mlsanity.com/paper/2109.00486/) (2021)
- [Improving Multilingual Translation by Representation and Gradient Regularization](https://mlsanity.com/paper/2109.04778/) (2021)
- [How to Design Translation Prompts for ChatGPT: An Empirical Study](https://mlsanity.com/paper/2304.02182/) (2023)
- [TLP: A Deep Learning-based Cost Model for Tensor Program Tuning](https://mlsanity.com/paper/2211.03578/) (2022)
- [Neural Machine Translation for the Indigenous Languages of the Americas: An Introduction](https://mlsanity.com/paper/2306.06804/) (2023)
- [AsPOS: Assamese Part of Speech Tagger using Deep Learning Approach](https://mlsanity.com/paper/2212.07043/) (2022)
- [Part-of-Speech Tagger for Bodo Language using Deep Learning approach](https://mlsanity.com/paper/2401.03175/) (2024)
- [Context-Aware Machine Translation with Source Coreference Explanation](https://mlsanity.com/paper/2404.19505/) (2024)
- [PHRASED: Phrase Dictionary Biasing for Speech Translation](https://mlsanity.com/paper/2506.09175/) (2025)
- [MetaXT: Meta Cross-Task Transfer between Disparate Label Spaces](https://mlsanity.com/paper/2109.04240/) (2021)
- [Multitask Finetuning for Improving Neural Machine Translation in Indian Languages](https://mlsanity.com/paper/2112.01742/) (2021)

## Abstract (arXiv)

Linguistic resources such as part-of-speech (POS) tags have been extensively used in statistical machine translation (SMT) frameworks and have yielded better performances. However, usage of such linguistic annotations in neural machine translation (NMT) systems has been left under-explored. In this work, we show that multi-task learning is a successful and a easy approach to introduce an additional knowledge into an end-to-end neural attentional model. By jointly training several natural language processing (NLP) tasks in one system, we are able to leverage common information and improve the performance of the individual task. We analyze the impact of three design decisions in multi-task learning: the tasks used in training, the training schedule, and the degree of parameter sharing across the tasks, which is defined by the network architecture. The experiments are conducted for an German to English translation task. As additional linguistic resources, we exploit POS information and named-entities (NE). Experiments show that the translation quality can be improved by up to 1.5 BLEU points under the low-resource condition. The performance of the POS tagger is also improved using the multi-task learning scheme.