AI RESEARCH

Frequency Autoregressive Image Generation with Continuous Tokens

arXiv CS.CV

ArXi:2503.05305v2 Announce Type: replace Autoregressive (AR) models for image generation typically adopt a two-stage paradigm of vector quantization and raster-scan ``next-token prediction", inspired by its great success in language modeling. However, due to the huge modality gap, image autoregressive models may require a systematic reevaluation from two perspectives: tokenizer format and regression direction. In this paper, we