Revisiting Autoregressive Models for Generative Image Classification

ArXi:2603.19122v1 Announce Type: new Class-conditional generative models have emerged as accurate and robust classifiers, with diffusion models nstrating clear advantages over other visual generative paradigms, including autoregressive (AR) models. In this work, we revisit visual AR-based generative classifiers and identify an important limitation of prior approaches: their reliance on a fixed token order, which imposes a restrictive inductive bias for image understanding.