What Makes Vocabulary Difficult?

The plots for each word explain the vocabulary difficulty predictions of the explainable model by team 🌸 Sakura at BEA 2026 Shared Task on Vocabulary Difficulty Prediction (arXiv paper and GitHub repo). The plotted SHAP values indicate the contribution of each group of features (e.g. L1 Similarity) to the specific prediction.

L1 Chinese (CN)

Chinese grouped SHAP plot

L1 German (DE)

German grouped SHAP plot

L1 Spanish (ES)

Spanish grouped SHAP plot