# ALIGN-MLM: Word Embedding Alignment is Crucial for Multilingual Pre-training

arXiv: https://arxiv.org/abs/2211.08547 · HTML: https://mlsanity.com/paper/2211.08547/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 29 of its 35 references that are in the map.
- First posted on arXiv 15 Nov 2022.

## Builds on (most cited first, 29 of 29)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [Pointer Sentinel Mixture Models](https://mlsanity.com/paper/1609.07843/) (2016)
- [Neural Machine Translation of Rare Words with Subword Units](https://mlsanity.com/paper/1508.07909/) (2015)
- [mT5: A massively multilingual pre-trained text-to-text transformer](https://mlsanity.com/paper/2010.11934/) (2020)
- [Cross-lingual Language Model Pretraining](https://mlsanity.com/paper/1901.07291/) (2019)
- [XNLI: Evaluating Cross-lingual Sentence Representations](https://mlsanity.com/paper/1809.05053/) (2018)
- [XTREME: A Massively Multilingual Multi-task Benchmark for Evaluating Cross-lingual Generalization](https://mlsanity.com/paper/2003.11080/) (2020)
- [How multilingual is Multilingual BERT?](https://mlsanity.com/paper/1906.01502/) (2019)
- [Word Translation Without Parallel Data](https://mlsanity.com/paper/1710.04087/) (2017)
- [Beto, Bentz, Becas: The Surprising Cross-Lingual Effectiveness of BERT](https://mlsanity.com/paper/1904.09077/) (2019)
- [Word Alignment by Fine-tuning Embeddings on Parallel Corpora](https://mlsanity.com/paper/2101.08231/) (2021)
- [Cross-Lingual Ability of Multilingual BERT: An Empirical Study](https://mlsanity.com/paper/1912.07840/) (2019)
- [Emerging Cross-lingual Structure in Pretrained Language Models](https://mlsanity.com/paper/1911.01464/) (2019)
- [MuRIL: Multilingual Representations for Indian Languages](https://mlsanity.com/paper/2103.10730/) (2021)
- [Choosing Transfer Languages for Cross-Lingual Learning](https://mlsanity.com/paper/1905.12688/) (2019)
- [A Survey Of Cross-lingual Word Embedding Models](https://mlsanity.com/paper/1706.04902/) (2017)
- [Multilingual Alignment of Contextual Word Representations](https://mlsanity.com/paper/2002.03518/) (2020)
- [Cross-Lingual Alignment of Contextual Word Embeddings, with Applications to Zero-shot Dependency Parsing](https://mlsanity.com/paper/1902.09492/) (2019)
- [Cross-Lingual BERT Transformation for Zero-Shot Dependency Parsing](https://mlsanity.com/paper/1909.06775/) (2019)
- [When is BERT Multilingual? Isolating Crucial Ingredients for Cross-lingual Transfer](https://mlsanity.com/paper/2110.14782/) (2021)
- [DICT-MLM: Improved Multilingual Pre-Training using Bilingual Dictionaries](https://mlsanity.com/paper/2010.12566/) (2020)
- [The Galactic Dependencies Treebanks: Getting More Data by Synthesizing New Languages](https://mlsanity.com/paper/1710.03838/) (2017)
- [Exploiting Language Relatedness for Low Web-Resource Language Model Adaptation: An Indic Languages Study](https://mlsanity.com/paper/2106.03958/) (2021)
- [Zero-shot Reading Comprehension by Cross-lingual Transfer Learning with Multi-lingual Language Representation Model](https://mlsanity.com/paper/1909.09587/) (2019)
- [Oolong: Investigating What Makes Transfer Learning Hard with Controlled Studies](https://mlsanity.com/paper/2202.12312/) (2022)
- [Bilingual Alignment Pre-Training for Zero-Shot Cross-Lingual Transfer](https://mlsanity.com/paper/2106.01732/) (2021)

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

- [Understanding Cross-Lingual Alignment -- A Survey](https://mlsanity.com/paper/2404.06228/) (2024)
- [PreAlign: Boosting Cross-Lingual Transfer by Early Establishment of Multilingual Alignment](https://mlsanity.com/paper/2407.16222/) (2024)
- [Breaking the Script Barrier in Multilingual Pre-Trained Language Models with Transliteration-Based Post-Training Alignment](https://mlsanity.com/paper/2406.19759/) (2024)
- [AlignX: Advancing Multilingual Large Language Models with Multilingual Representation Alignment](https://mlsanity.com/paper/2509.24338/) (2025)
- [Bridging Linguistic Gaps: Cross-Lingual Mapping in Pre-Training and Dataset for Enhanced Multilingual LLM Performance](https://mlsanity.com/paper/2604.10590/) (2026)
- [Enhancing Multilingual Embeddings via Multi-Way Parallel Text Alignment](https://mlsanity.com/paper/2602.21543/) (2026)

## Abstract (arXiv)

Multilingual pre-trained models exhibit zero-shot cross-lingual transfer, where a model fine-tuned on a source language achieves surprisingly good performance on a target language. While studies have attempted to understand transfer, they focus only on MLM, and the large number of differences between natural languages makes it hard to disentangle the importance of different properties. In this work, we specifically highlight the importance of word embedding alignment by proposing a pre-training objective (ALIGN-MLM) whose auxiliary loss guides similar words in different languages to have similar word embeddings. ALIGN-MLM either outperforms or matches three widely adopted objectives (MLM, XLM, DICT-MLM) when we evaluate transfer between pairs of natural languages and their counterparts created by systematically modifying specific properties like the script. In particular, ALIGN-MLM outperforms XLM and MLM by 35 and 30 F1 points on POS-tagging for transfer between languages that differ both in their script and word order (left-to-right v.s. right-to-left). We also show a strong correlation between alignment and transfer for all objectives (e.g., rho=0.727 for XNLI), which together with ALIGN-MLM's strong performance calls for explicitly aligning word embeddings for multilingual models.