Skip to content
QbitAI · WeChat

ByteDance Proposes Generative Refinement Networks as a Third Route for Visual Generation

挑战扩散自回归统治!字节提出视觉生成第三种路线,让模型像人类一样边画边改

ByteDance’s commercial technology team proposed GRN, a visual generation architecture using HBQ, global refinement, and complexity-aware sampling to address quantization loss, error accumulation, and fixed-step inference; on a 130M model, adaptive sampling reduced inference from 50 steps to an average of 24, while gFID changed from 3.56 to 3.79.

Why it matters: ByteDance's GRN drafts a coarse image then refines it, with HBQ layered binary quantization cutting loss in discrete codes, a concrete third path beyond diffusion and autoregression.

Read the original ↗Export Markdown