SOURCE-LINKED INTELLIGENCE

Prefix-Denoising Consistency: Test-Time Verification for Diffusion Language Models

arXiv · AI, language, vision and robotics · article · Aug 26, 2026 · UTC

Diffusion Language Models (DLMs) have recently become increasingly competitive with autoregressive (AR) models, and even outperform them on certain tasks. Unlike AR models, DLMs produce output through iterative denoising without a left-to-right order. To further improve the performance of DLMs, we introduce PDC (\emph{Prefix-Denoising Consistency}), a test-time self-verification method for DLMs. PDC exploits a distinctive test-time signal in DLMs under prefix conditioned regeneration, correct trajectories are more stable and reproducible than incorrect ones. Concretely, given an initially gene

Read original source ↗ Open in workspace

recordType

paper

region

Global