# Multilingual Factor Analysis

## TL;DR

This paper introduces a generative latent variable model for learning multilingual word representations from dictionaries, enabling robust alignment across languages and performing well despite noisy data.

## Contribution

It presents a novel offline approach using a generative model to align multilingual embeddings, improving robustness and alignment quality.

## Key findings

- Achieves competitive results on multilingual tasks
- Robust to noise in embedding space
- Effective for distributed representations from noisy corpora

## Abstract

In this work we approach the task of learning multilingual word representations in an offline manner by fitting a generative latent variable model to a multilingual dictionary. We model equivalent words in different languages as different views of the same word generated by a common latent variable representing their latent lexical meaning. We explore the task of alignment by querying the fitted model for multilingual embeddings achieving competitive results across a variety of tasks. The proposed model is robust to noise in the embedding space making it a suitable method for distributed representations learned from noisy corpora.