# Finding Universal Grammatical Relations in Multilingual BERT

arXiv: https://arxiv.org/abs/2005.04511 · HTML: https://mlsanity.com/paper/2005.04511/

## Key facts

- Cited by 110 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 May 2020.

## Built on by (most cited first, 30 of 110)

- [Large Language Models](https://mlsanity.com/paper/2307.05782/) (2023)
- [Measuring and Improving Consistency in Pretrained Language Models](https://mlsanity.com/paper/2102.01017/) (2021)
- [Causal Analysis of Syntactic Agreement Mechanisms in Neural Language Models](https://mlsanity.com/paper/2106.06087/) (2021)
- [Towards Faithful Model Explanation in NLP: A Survey](https://mlsanity.com/paper/2209.11326/) (2022)
- [The Interplay of Variant, Size, and Task Type in Arabic Pre-trained Language Models](https://mlsanity.com/paper/2103.06678/) (2021)
- [Interpretability at Scale: Identifying Causal Mechanisms in Alpaca](https://mlsanity.com/paper/2305.08809/) (2023)
- [The Geometry of Multilingual Language Model Representations](https://mlsanity.com/paper/2205.10964/) (2022)
- [LLaMA Beyond English: An Empirical Study on Language Capability Transfer](https://mlsanity.com/paper/2401.01055/) (2024)
- [MultiBLiMP 1.0: A Massively Multilingual Benchmark of Linguistic Minimal Pairs](https://mlsanity.com/paper/2504.02768/) (2025)
- [It's All in the Heads: Using Attention Heads as a Baseline for Cross-Lingual Transfer in Commonsense Reasoning](https://mlsanity.com/paper/2106.12066/) (2021)
- [First Align, then Predict: Understanding the Cross-Lingual Ability of Multilingual BERT](https://mlsanity.com/paper/2101.11109/) (2021)
- [Codified audio language modeling learns useful representations for music information retrieval](https://mlsanity.com/paper/2107.05677/) (2021)
- [Large Language Models Share Representations of Latent Grammatical Concepts Across Typologically Diverse Languages](https://mlsanity.com/paper/2501.06346/) (2025)
- [AttentionViz: A Global View of Transformer Attention](https://mlsanity.com/paper/2305.03210/) (2023)
- [A Primer on Pretrained Multilingual Language Models](https://mlsanity.com/paper/2107.00676/) (2021)
- [Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers](https://mlsanity.com/paper/2411.08745/) (2024)
- [Deep Subjecthood: Higher-Order Grammatical Features in Multilingual BERT](https://mlsanity.com/paper/2101.11043/) (2021)
- [Models and Datasets for Cross-Lingual Summarisation](https://mlsanity.com/paper/2202.09583/) (2022)
- [Lost in Translation: Large Language Models in Non-English Content Analysis](https://mlsanity.com/paper/2306.07377/) (2023)
- [Analyzing the Mono- and Cross-Lingual Pretraining Dynamics of Multilingual Language Models](https://mlsanity.com/paper/2205.11758/) (2022)
- [Exploring the Role of BERT Token Representations to Explain Sentence Probing Results](https://mlsanity.com/paper/2104.01477/) (2021)
- [Goldfish: Monolingual Language Models for 350 Languages](https://mlsanity.com/paper/2408.10441/) (2024)
- [Coloring the Blank Slate: Pre-training Imparts a Hierarchical Inductive Bias to Sequence-to-sequence Models](https://mlsanity.com/paper/2203.09397/) (2022)
- [Pretraining with Artificial Language: Studying Transferable Knowledge in Language Models](https://mlsanity.com/paper/2203.10326/) (2022)
- [Discovering Language-neutral Sub-networks in Multilingual Language Models](https://mlsanity.com/paper/2205.12672/) (2022)
- [Insights into Pre-training via Simpler Synthetic Tasks](https://mlsanity.com/paper/2206.10139/) (2022)
- [An Isotropy Analysis in the Multilingual BERT Embedding Space](https://mlsanity.com/paper/2110.04504/) (2021)
- [A polar coordinate system represents syntax in large language models](https://mlsanity.com/paper/2412.05571/) (2024)
- [On the Similarity of Circuits across Languages: a Case Study on the Subject-verb Agreement Task](https://mlsanity.com/paper/2410.06496/) (2024)
- [Relational Composition in Neural Networks: A Survey and Call to Action](https://mlsanity.com/paper/2407.14662/) (2024)

## Abstract (arXiv)

Recent work has found evidence that Multilingual BERT (mBERT), a transformer-based multilingual masked language model, is capable of zero-shot cross-lingual transfer, suggesting that some aspects of its representations are shared cross-lingually. To better understand this overlap, we extend recent work on finding syntactic trees in neural networks' internal representations to the multilingual setting. We show that subspaces of mBERT representations recover syntactic tree distances in languages other than English, and that these subspaces are approximately shared across languages. Motivated by these results, we present an unsupervised analysis method that provides evidence mBERT learns representations of syntactic dependency labels, in the form of clusters which largely agree with the Universal Dependencies taxonomy. This evidence suggests that even without explicit supervision, multilingual masked language models learn certain linguistic universals.