Improving Multilingual Language Models by Aligning Representations through Steering

Omar Mohamed Mahmoud, Buddhika Laknath Semage, Thommen George Karimpanal, Santu Rana


Abstract
This paper investigates how Large Language Models (LLMs) represent non-English tokens—a question that remains underexplored despite recent progress. We propose a lightweight intervention method using representation steering, where a learned vector is added to the residual stream at a single model layer to enhance multilingual performance. Through extensive experiments across seven competitive baselines—including prompt optimization, supervised fine-tuning (SFT), in-context learning, cross-lingual transfer, projection mapping techniques, and translation-based methods—we show that our approach consistently outperforms most alternatives. In particular, it achieves performance on par with production-grade translation systems while requiring far fewer resources. We further explore the complementarity between our method and SFT, demonstrating that steering offers a direct, efficient way to realign internal representations. These findings underscore the potential of activation-level interventions as a powerful tool for improving the multilingual capabilities of LLMs.
Anthology ID:
2026.lrec-1.164
Volume:
Proceedings of the Fifteenth Language Resources and Evaluation Conference
Month:
May
Year:
2026
Address:
Palma de Mallorca, Spain
Editors:
Stelios Piperidis, Núria Bel, Henk van den Heuvel, Nancy Ide, Simon Krek, Antonio Toral
Venue:
LREC
SIG:
Publisher:
ELRA Language Resource Association
Note:
Pages:
2090–2103
Language:
External URL:
https://lrec.elra.info/lrec2026-main-164
DOI:
10.63317/244dsoue8zu2
Bibkey:
Cite (ACL):
Omar Mohamed Mahmoud, Buddhika Laknath Semage, Thommen George Karimpanal, and Santu Rana. 2026. Improving Multilingual Language Models by Aligning Representations through Steering. In Proceedings of the Fifteenth Language Resources and Evaluation Conference, pages 2090–2103, Palma de Mallorca, Spain. ELRA Language Resource Association.
Cite (Informal):
Improving Multilingual Language Models by Aligning Representations through Steering (Mahmoud et al., LREC 2026)
Copy Citation: