Crosslinguistic Semantic Textual Similarity of Buddhist Chinese and Classical Tibetan

In this paper we present the first-ever procedure for identifying highly similar sequences of text in Chinese and Tibetan translations of Buddhist sutra literature. We initially propose this procedure as an aid to scholars engaged in the philological study of Buddhist documents. We create a cross-lingual embedding space by taking the cosine similarity of average sequence vectors in order to produce unsupervised similar cross-linguistic parallel alignments at word, sentence, and even paragraph level. Initial results show that our method lays a solid foundation for the future development of a fully-fledged Information Retrieval tool for these (and potentially other) low-resource historical languages.