Disentangling Language Roles in Multilingual LLM Task Execution 事件

Disentangling Language Roles in Multilingual LLM Task Execution arXiv:2605.27649v1 Announce Type: new Abstract: 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 introduce MTM-Bench, a controlled benchmark for language-conditioned task execution in which each instance