We present the largest multilingual experiment to date on word-to-morph segmentation, covering 58 typologically diverse languages. We describe a newly compiled collection of linguistically annotated resources for the task, providing broad coverage and enabling systematic cross-lingual evaluation. Second, we train two neural models on surface morphological segmentation, achieving 81% average word accuracy on the original datasets, slightly outperforming previous methods. Experiments on custom test sets reveal substantial variation in performance, highlighting the need for further harmonization and more robust multilingual approaches.

This study evaluates the extent to which semantic information is preserved within sentence embeddings generated from state-of-art sentence embedding models: SBERT and LaBSE. Specifically, we analyzed 13 semantic attributes in sentence embeddings. Our findings indicate that some semantic features (such as tense-related classes) can be decoded from the representation of sentence embeddings. Additionally, we discover the limitation of the current sentence embedding models: inferring meaning beyond the lexical level has proven to be difficult.