AI RESEARCH
Path-Dependent Denoising: A Non-Conservative Field Perspective on Order Collapse in Diffusion Language Models
arXiv CS.LG
•
ArXi:2605.09303v1 Announce Type: new Diffusion language models (DLMs) offer a structural alternative to autoregressive generation: denoising can update tokens in arbitrary orders or in parallel rather than along a fixed left-to-right chain. In practice, fast DLM decoding remains strongly order-sensitive and often drifts toward autoregressive-like trajectories. We trace this tension to compatibility. At each reverse-time step, a DLM provides local denoising conditionals over the unresolved tokens.