Subsets of WMT13 Parallel Data for Evaluating Cross-Lingual Embeddings

Random subsets of WMT'13 Spanish-English and German-English parallel dataset (common crawl) used to evaluate cross-lingual sentence embeddings as described in "Context-Aware Crosslingual Mapping"

本数据集为WMT'13西班牙语-英语、德语-英语平行数据集（源自通用爬虫（Common Crawl））的随机子集，用于按照《上下文感知跨语言映射》（Context-Aware Crosslingual Mapping）中所述方案评估跨语言句子嵌入（cross-lingual sentence embeddings）。