K-means clustering

Partition vectors into clusters by iteratively assigning each to the nearest centroid and recomputing centroids.

Findings

• IC-009LLM moral preferences show significant language sensitivity but not inequality toward low-resource languages [primary]
• IC-667CLIP's intermediate layer features encode object boundaries recoverable by k-means clustering, a property absent in shallow and deep layers [primary]
• IC-669DINOv2's features, when clustered, produce smooth semantic regions but lack instance-level boundary delineation [primary]
• IC-673Trained depthwise convolutional kernels in DS-CNN architectures converge to identifiable DoG-like patterns, with over 95% of ConvNeXtV2 and over 90% of ConvNeXt filters classifiable into a small set of clusters [supporting]