标签
本研究评估了多语言句子嵌入在区分正确英希翻译与错误翻译方面的表现,发现这些嵌入提供了有用的语义信号,但更适合集成到更广泛的翻译评估框架中。
本研究使用多语言编码器来测量Wikipedia各语言版本间的框架差异,发现宗教和政治概念的分歧大于科学概念。
This paper introduces Task-Conditional Flow Matching (TCFM), a framework for adapting multilingual text embedding models that selectively uses flow matching for translation tasks and other objectives for retrieval/classification. It achieves state-of-the-art results on the Indic Massive Text Embedding Benchmark.