Aligned representations across languages is a desired property in multilingual large language models (mLLMs), as alignment can improve performance in cross-lingual tasks. Typically alignment requires fine-tuning a model, which is computationally expensive, and sizable language data, which often may not be available. A data-efficient alternative to fine-tuning is model interventions -- a method for manipulating model activations to steer generation into the desired direction. We analyze the effect of a popular intervention (finding experts) on the alignment of cross-lingual representations in mLLMs. We identify the neurons to manipulate for a given language and introspect the embedding space of mLLMs pre- and post-manipulation. We show that modifying the mLLM's activations changes its embedding space such that cross-lingual alignment is enhanced. Further, we show that the changes to the embedding space translate into improved downstream performance on retrieval tasks, with up to 2x improvements in top-1 accuracy on cross-lingual retrieval.
View on arXiv@article{sundar2025_2502.15639,
title={ Steering into New Embedding Spaces: Analyzing Cross-Lingual Alignment Induced by Model Interventions in Multilingual Language Models },
author={ Anirudh Sundar and Sinead Williamson and Katherine Metcalf and Barry-John Theobald and Skyler Seto and Masha Fedzechkina },
journal={arXiv preprint arXiv:2502.15639},
year={ 2025 }
}