AI RESEARCH

Disentangling Language Roles in Multilingual LLM Task Execution

arXiv CS.LG

ArXi:2605.27649v1 Announce Type: cross Multilingual LLMs are increasingly used when instruction, source content, and required response languages do not coincide. Existing benchmarks have expanded multilingual instruction-following evaluation, but they rarely isolate these three roles within a fully crossed design. We