We presented two posters at CogSci 2023. Evaluating machine comprehension of sketch meaning at different levels of abstraction and Behavioral estimates of conceptual structure are robust across tasks in humans but not large language models. See publications for more details~