Jelke Bloem

Investigating How LLMs Propagate Female Stereotypes: Comparing What Models Say via Prompts with What They Represent in Their Embeddings

Multi-SimLex for Dutch: Benchmarking Embedding- and Prompt-Based Model Performance on Semantic Similarity

Dutch Multi-SimLex is introduced, a 1,888–pair extension of the Multi-SimLex benchmark for evaluating lexical semantic similarity in Dutch, and results show that model performance depends strongly on how meaning is tested.