Tag
This paper compares semantic search dynamics between humans and LLMs using verbal fluency data, finding that humans exhibit more variable and exploratory search patterns that current models fail to reproduce.
This paper systematically studies how different evaluation objectives (accuracy, silhouette score, PCA reconstruction loss) and subset-size regularization directions affect search dynamics and solution quality in multiobjective unsupervised feature selection, showing that silhouette-based formulations bias toward trivial low-cardinality solutions while PCA loss yields compact subsets with competitive accuracy.