Steering beyond Language: Improving Multilingual Training-Free Text Embeddings via Paired-Residual Subspace Removal

Paper recorded by Signals 4 on 2026-10-16 in cs.CL. Abstract reproduced from arXiv; link to the original below.

Published 2026-10-16 on arXiv · recorded by Signals 4 on 2026-10-19

Category: cs.CL · 自然语言处理 · first seen 2026-10-19

Abstract

While deriving text embeddings directly from frozen LLMs has emerged as a promising direction, existing methods focus almost exclusively on English, leaving multilingual settings largely unexplored. In this work, we present the first systematic study of multilingual training-free text embeddings in LLMs. Our analysis reveals a language-conditioned embedding geometry in which semantically equivalen

Read on arXiv →