In a comparison setting (two stimuli), we tested 4- and 6-year-old children’s generalization of novel names for objects. We manipulated the semantic distance between the two training items (e.g., two bracelets versus a bracelet and a watch), and the semantic distance between the training items and the test items (e.g., a pendant versus a bow tie). We tested whether smaller semantic distance between training items would lead to more taxonomic (vs. perceptual) choices at test, than broader semantic distance during training, especially in the case of distant test stimuli. Results revealed main effects of induction distance, of generalization distance and that only children aged 6 years benefited from broader semantic during training at test. Four year-old children failed to generalize to far test stimuli even with semantically distant training items. We discuss how conceptual distance during learning differentially affects generalization performance across age groups.