paper-with-me

Papers

Improving Multilingual Language Models by Aligning Representations through Steering

2025-05-19 · Omar Mahmoud, Buddhika Laknath Semage, Thommen George Karimpanal, Santu Rana

In this paper, we investigate how large language models (LLMS) process non-English tokens within their layer representations, an open question despite significant advancements in the field. Using representation steering, specifically by adding a learned vector to a single model layer's activations, we demonstrate that steering a single model layer can notably enhance performance. Our analysis shows that this approach achieves results comparable to translation baselines and surpasses state of the art prompt optimization methods. Additionally, we highlight how advanced techniques like supervised fine tuning (\textsc{sft}) and reinforcement learning from human feedback (\textsc{rlhf}) improve multilingual capabilities by altering representation spaces. We further illustrate how these methods align with our approach to reshaping LLMS layer representations.

📄 PDF Abstract BibTeX arXiv:2505.12584

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

ALIGN In the ALIGN method, visual and language representations are jointly trained from noisy image alt-text data. The image and text encoders are learned via contrastive loss…

Similar Papers 제목 키워드 기반

Bias Beyond Borders: Political Ideology Evaluation and Steering in Multilingual LLMs

2026-01-30 · Afrozah Nadeem, Agrima Seth, Mehwish Nasim, Usman Naseem arxiv

Large Language Models (LLMs) increasingly shape global discourse, making fairness and ideological neutrality essential for responsible AI deployment. Despite growing attention to political bias in LLMs, prior work largel…

Steering Risk Preferences in Large Language Models by Aligning Behavioral and Neural Representations

2025-05-16 · Jian-Qiao Zhu, Haijiang Yan, Thomas L. Griffiths

Changing the behavior of large language models (LLMs) can be as straightforward as editing the Transformer's residual streams using appropriately constructed "steering vectors." These modifications to internal neural act…

Cross-Lingual Activation Steering for Multilingual Language Models

2026-01-23 · Rhitabrat Pokharel, Ameeta Agrawal, Tanay Nagar arxiv

Large language models exhibit strong multilingual capabilities, yet significant performance gaps persist between dominant and non-dominant languages. Prior work attributes this gap to imbalances between shared and langua…

Multilingual Steering by Design: Multilingual Sparse Autoencoders and Principled Layer Selection

2026-05-21 · Yusser Al Ghussin, Daniil Gurgurov, Tanja Baeumel, Josef van Genabith 외 arxiv

Sparse autoencoders (SAEs) enable feature-level mechanistic interpretability and activation steering in large language models (LLMs), but SAE-based language control remains unreliable in multilingual settings: most SAEs …

Language IdentificationMachine Translation

CLaS-Bench: A Cross-Lingual Alignment and Steering Benchmark

2026-01-13 · Daniil Gurgurov, Yusser Al Ghussin, Tanja Baeumel, Cheng-Ting Chou 외 arxiv

Understanding and controlling the behavior of large language models (LLMs) is an increasingly important topic in multilingual NLP. Beyond prompting or fine-tuning, , i.e.,~manipulating internal representations during inf…