SOURCE-LINKED INTELLIGENCE

Diffusion as a Training Curriculum for Timestep-Free Iterative Reasoning

arXiv · AI, language, vision and robotics · article · Sep 1, 2026 · UTC

Diffusion models and recursive reasoners are both iterative, but they carry information across iterations differently. We add a persistent hidden state to a diffusion denoiser and remove its timestep conditioning, leaving a single shared update that can be run to arbitrary depth. The result is an anytime solver: accuracy keeps improving with inference depth far beyond the rollout lengths and backpropagation window used in training, reaching 99.90% exact solve on Sudoku-Extreme. We also obtain 98.93% solve rate on Maze-Unique. Surprisingly, progressive denoising is unnecessary at inference: hol

Read original source ↗ Open in workspace

recordType

paper

region

Global