# TAMS: Translation-Assisted Morphological Segmentation

arXiv: https://arxiv.org/abs/2403.14840 · HTML: https://mlsanity.com/paper/2403.14840/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 31 references that are in the map.
- First posted on arXiv 21 Mar 2024.

## Builds on (most cited first, 10 of 10)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Neural Machine Translation by Jointly Learning to Align and Translate](https://mlsanity.com/paper/1409.0473/) (2014)
- [Get To The Point: Summarization with Pointer-Generator Networks](https://mlsanity.com/paper/1704.04368/) (2017)
- [Pointer Networks](https://mlsanity.com/paper/1506.03134/) (2015)
- [Are All Languages Created Equal in Multilingual BERT?](https://mlsanity.com/paper/2005.09093/) (2020)
- [Word Alignment by Fine-tuning Embeddings on Parallel Corpora](https://mlsanity.com/paper/2101.08231/) (2021)
- [Small and Practical BERT Models for Sequence Labeling](https://mlsanity.com/paper/1909.00100/) (2019)
- [Tackling the Low-resource Challenge for Canonical Segmentation](https://mlsanity.com/paper/2010.02804/) (2020)
- [Canonical and Surface Morphological Segmentation for Nguni Languages](https://mlsanity.com/paper/2104.00767/) (2021)
- [AUTOLEX: An Automatic Framework for Linguistic Exploration](https://mlsanity.com/paper/2203.13901/) (2022)

## Abstract (arXiv)

Canonical morphological segmentation is the process of analyzing words into the standard (aka underlying) forms of their constituent morphemes. This is a core task in language documentation, and NLP systems have the potential to dramatically speed up this process. But in typical language documentation settings, training data for canonical morpheme segmentation is scarce, making it difficult to train high quality models. However, translation data is often much more abundant, and, in this work, we present a method that attempts to leverage this data in the canonical segmentation task. We propose a character-level sequence-to-sequence model that incorporates representations of translations obtained from pretrained high-resource monolingual language models as an additional signal. Our model outperforms the baseline in a super-low resource setting but yields mixed results on training splits with more data. While further work is needed to make translations useful in higher-resource settings, our model shows promise in severely resource-constrained settings.