We focus on automatically evaluating contextual informativeness relative to multiple target words in child-directed text, with implications for improving the automatic generation of educational stories for early childhood vocabulary intervention. Can’t wait to share, and learn about others’ work :)