paper-with-me

Papers

Steering into New Embedding Spaces: Analyzing Cross-Lingual Alignment Induced by Model Interventions in Multilingual Language Models

2025-02-21 · Anirudh Sundar, Sinead Williamson, Katherine Metcalf, Barry-John Theobald, Skyler Seto, Masha Fedzechkina

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.

📄 PDF Abstract BibTeX arXiv:2502.15639

Code (0)

등록된 구현이 없습니다.

Tasks

Retrieval

Similar Papers 제목 키워드 기반

LLM-Augmented Semantic Steering of Text Embedding Projection Spaces

2026-05-03 · Wei Liu, Eric Krokos, Kirsten Whitley, Rebecca Faust 외 arxiv

Low-dimensional projections of text embeddings support visual analysis of document collections, but their spatial organization may not reflect the relationships an analyst intends to examine. Existing semantic interactio…

Curveball Steering: The Right Direction To Steer Isn't Always Linear

2026-03-10 · Shivam Raval, Hae Jin Song, Linlin Wu, Abir Harrasse 외 arxiv

Activation steering is a widely used approach for controlling large language model (LLM) behavior by intervening on internal representations. Existing methods largely rely on the Linear Representation Hypothesis, assumin…

DISCO: Disentangled Communication Steering for Large Language Models

2025-09-20 · Max Torop, Aria Masoomi, Masih Eskandar, Jennifer Dy arxiv

A variety of recent methods guide large language model outputs via the inference-time addition of steering vectors to residual-stream or attention-head representations. In contrast, we propose to inject steering vectors …

Emblaze: Illuminating Machine Learning Representations through Interactive Comparison of Embedding Spaces

2022-02-05 · Venkatesh Sivaraman, Yiwei Wu, Adam Perer

Modern machine learning techniques commonly rely on complex, high-dimensional embedding representations to capture underlying structure in the data and improve performance. In order to characterize model flaws and choose…

BIG-bench Machine Learning

Spectral Attention Steering for Prompt Highlighting

2026-03-01 · Weixian Waylon Li, Yuchen Niu, Yongxin Yang, Keshuang Li 외 arxiv

Attention steering is an important technique for controlling model focus, enabling capabilities such as prompt highlighting, where the model prioritises user-specified text. However, existing attention steering methods r…