标签
This paper presents a cost-efficient routing pipeline for multilingual short-text classification that selectively translates low-resource languages into English before zero-shot classification using small language models, showing quality gains on SIB-200 and MASSIVE benchmarks.