AI RESEARCH
Beyond Bilingual Transfer: Multilingual Code-Switching in Instruction Tuning
arXiv CS.AI
•
ArXi:2605.29414v1 Announce Type: cross Recent studies have shown that code-switching data (CSD), in which multiple languages are mixed within the same context, can improve cross-lingual transfer and multilingual alignment in large language models (LLMs). However, existing studies primarily focus on bilingual transfer between English and a target language, leaving multilingual settings involving three or languages largely unexplored. In this work, we investigate multilingual code-switching instruction tuning across four languages: English, Japanese, Korean, and Chinese.