# Examining the Limits of Word2Vec with Toki Pona

arXiv: https://arxiv.org/abs/2606.17299 · HTML: https://mlsanity.com/paper/2606.17299/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 5 of its 15 references that are in the map.
- First posted on arXiv 15 Jun 2026.

## Builds on (most cited first, 5 of 5)

- [Efficient Estimation of Word Representations in Vector Space](https://mlsanity.com/paper/1301.3781/) (2013)
- [UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction](https://mlsanity.com/paper/1802.03426/) (2018)
- [SimLex-999: Evaluating Semantic Models with (Genuine) Similarity Estimation](https://mlsanity.com/paper/1408.3456/) (2014)
- [On the Dimensionality of Word Embedding](https://mlsanity.com/paper/1812.04224/) (2018)
- [Semantic Drift in Multilingual Representations](https://mlsanity.com/paper/1904.10820/) (2019)

## Abstract (arXiv)

Word2Vec's effectiveness at generating semantic embeddings has been widely validated, yet it has been tested almost exclusively on languages with large vocabulary inventories. This study examines whether Word2Vec can successfully capture semantic relationships within an extremely reduced vocabulary using data from Toki Pona, a constructed language with approximately 130 words. We sourced 1.4 million sentences (7.95 million tokens) from the Toki Pona community for training. Approximately 23% of sentences in the corpus contain non-Toki Pona tokens such as named entities, loanwords, and neologisms. To investigate whether this linguistic noise enhances or hinders performance -- a topic rarely addressed in word embedding literature -- we trained two distinct models: one retaining these incidental tokens and another filtering them out completely. Evaluation was conducted using quantitative methods measuring word proximity to semantic category centroids, automated silhouette scores via agglomerative clustering, and qualitative analysis utilizing representational similarity matrices compared against English. The results indicate that while sparse, non-core tokens do not affect the relative structure of the learned embeddings, they actually draw similar words closer together in the vector space. Importantly, Word2Vec's effectiveness depends more on distributional patterns than lexicon size even at this extreme lower bound.