AI RESEARCH
Language steering in latent space to mitigate unintended code-switching
arXiv CS.LG
•
ArXi:2510.13849v3 Announce Type: replace-cross Multilingual Large Language Models (LLMs) often exhibit hallucinations such as unintended code-switching, reducing reliability in downstream tasks. We propose latent-space language steering, a lightweight inference-time method that identifies language directions via Principal Component Analysis (PCA) on parallel translations and steers token embeddings along these axes to control language identity.