SyncAI.news, a Varaisys broadcasting
Beyond Bilingual Transfer: Multilingual Code-Switching in Instruction Tuning
SA

Shunta Asano, Jeonghun Baek, Toshihiko Yamasaki

· 1 min read

ResearcharXiv cs.CL

Beyond Bilingual Transfer: Multilingual Code-Switching in Instruction Tuning

arXiv:2605.29414v2 Announce Type: replace Abstract: 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). Recent studies on synthetic code-switching for LLMs have primarily focused on bilingual transfer between English and a target language, leaving instruction-tuning settings that jointly mix three or more languages within each example largely unexplored. In this work, we investigate multilingual code-switching instruction tuning across four languages: English, Japanese, Korean, and Chinese. We evaluate multilingual understanding on Belebele and MMLU. Our experiments show that simple sentence-level multilingual CSD consistently improves the four-language average accuracy across all bilingual, trilingual, and quadrilingual settings for Gemma-2-2B, indicating that multilingual code-switching can be effective beyond bilingual transfer settings.

Original source

This story was published by arXiv cs.CL and written by Shunta Asano, Jeonghun Baek, Toshihiko Yamasaki. SyncAI.news shows a preview; the complete article is on the publisher's site.

Read the full story on arxiv.org

Similar News