Skip to content
arXiv cs.CV · Papers

Anchored, Not Graded: Vision-Language Models Fail at Slant-from-Texture Perception

arXiv:2606.06714v2 Announce Type: replace Abstract: Human perception of surface slant from texture exhibits systematic, graded biases that emerge reliably in psychophysical experiments. Prior work showed that unsupervised CNNs reproduce several human-like biases, while supervised CNNs do not. Do Vision-Language Models