# On the Limitations of Unsupervised Bilingual Dictionary Induction

arXiv: https://arxiv.org/abs/1805.03620 · HTML: https://mlsanity.com/paper/1805.03620/

## Key facts

- Cited by 61 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 May 2018.

## Built on by (most cited first, 30 of 61)

- [Challenges and Strategies in Cross-Cultural NLP](https://mlsanity.com/paper/2203.10020/) (2022)
- [XTREME-R: Towards More Challenging and Nuanced Multilingual Evaluation](https://mlsanity.com/paper/2104.07412/) (2021)
- [Neural Machine Translation for Low-Resource Languages: A Survey](https://mlsanity.com/paper/2106.15115/) (2021)
- [A Survey on Multilingual Large Language Models: Corpora, Alignment, and Bias](https://mlsanity.com/paper/2404.00929/) (2024)
- [Understanding Cross-Lingual Alignment -- A Survey](https://mlsanity.com/paper/2404.06228/) (2024)
- [Do Vision and Language Models Share Concepts? A Vector Space Alignment Study](https://mlsanity.com/paper/2302.06555/) (2023)
- [Bilingual Lexicon Induction via Unsupervised Bitext Construction and Word Alignment](https://mlsanity.com/paper/2101.00148/) (2021)
- [Do Language Models Know the Way to Rome?](https://mlsanity.com/paper/2109.07971/) (2021)
- [Improving Word Translation via Two-Stage Contrastive Learning](https://mlsanity.com/paper/2203.08307/) (2022)
- [Transfer to a Low-Resource Language via Close Relatives: The Case Study on Faroese](https://mlsanity.com/paper/2304.08823/) (2023)
- [AM2iCo: Evaluating Word Meaning in Context across Low-Resource Languages with Adversarial Examples](https://mlsanity.com/paper/2104.08639/) (2021)
- [Itihasa: A large-scale corpus for Sanskrit to English translation](https://mlsanity.com/paper/2106.03269/) (2021)
- [Concept Space Alignment in Multilingual LLMs](https://mlsanity.com/paper/2410.01079/) (2024)
- [Combining Static Word Embeddings and Contextual Representations for Bilingual Lexicon Induction](https://mlsanity.com/paper/2106.03084/) (2021)
- [Subword Mapping and Anchoring across Languages](https://mlsanity.com/paper/2109.04556/) (2021)
- [Multi-source Learning via Completion of Block-wise Overlapping Noisy Matrices](https://mlsanity.com/paper/2105.10360/) (2021)
- [Beyond Shared Vocabulary: Increasing Representational Word Similarities across Languages for Multilingual Machine Translation](https://mlsanity.com/paper/2305.14189/) (2023)
- [Delving Deeper into Cross-lingual Visual Question Answering](https://mlsanity.com/paper/2202.07630/) (2022)
- [IsoVec: Controlling the Relative Isomorphism of Word Embedding Spaces](https://mlsanity.com/paper/2210.05098/) (2022)
- [Neural Machine Translation for the Indigenous Languages of the Americas: An Introduction](https://mlsanity.com/paper/2306.06804/) (2023)
- [Cultural Adaptation of Recipes](https://mlsanity.com/paper/2310.17353/) (2023)
- [Cross-Lingual BERT Contextual Embedding Space Mapping with Isotropic and Isometric Conditions](https://mlsanity.com/paper/2107.09186/) (2021)
- [Bilingual Lexicon Induction for Low-Resource Languages using Graph Matching via Optimal Transport](https://mlsanity.com/paper/2210.14378/) (2022)
- [The Impact of Cross-Lingual Adjustment of Contextual Word Representations on Zero-Shot Transfer](https://mlsanity.com/paper/2204.06457/) (2022)
- [indicnlp@kgp at DravidianLangTech-EACL2021: Offensive Language Identification in Dravidian Languages](https://mlsanity.com/paper/2102.07150/) (2021)
- [Did AI get more negative recently?](https://mlsanity.com/paper/2202.13610/) (2022)
- [The Shape of Word Embeddings: Quantifying Non-Isometry With Topological Data Analysis](https://mlsanity.com/paper/2404.00500/) (2024)
- [Towards Multi-Sense Cross-Lingual Alignment of Contextual Embeddings](https://mlsanity.com/paper/2103.06459/) (2021)
- [Cross-lingual alignments of ELMo contextual embeddings](https://mlsanity.com/paper/2106.15986/) (2021)
- [Systematic Investigation of Strategies Tailored for Low-Resource Settings for Low-Resource Dependency Parsing](https://mlsanity.com/paper/2201.11374/) (2022)

## Abstract (arXiv)

Unsupervised machine translation---i.e., not assuming any cross-lingual supervision signal, whether a dictionary, translations, or comparable corpora---seems impossible, but nevertheless, Lample et al. (2018) recently proposed a fully unsupervised machine translation (MT) model. The model relies heavily on an adversarial, unsupervised alignment of word embedding spaces for bilingual dictionary induction (Conneau et al., 2018), which we examine here. Our results identify the limitations of current unsupervised MT: unsupervised bilingual dictionary induction performs much worse on morphologically rich languages that are not dependent marking, when monolingual corpora from different domains or different embedding algorithms are used. We show that a simple trick, exploiting a weak supervision signal from identical words, enables more robust induction, and establish a near-perfect correlation between unsupervised bilingual dictionary induction performance and a previously unexplored graph similarity metric.