XIFBench: Evaluating Large Language Models on Multilingual Instruction Following | Read Paper on Bytez