Does Uniform Discrete Diffusion Need Time?

arXiv:2609.30977v1 Announce Type: cross Abstract: Uniform discrete diffusion models (UDMs) commonly use explicit time conditioning, but we find that it can often be unnecessary in practice. In this paper, we first show that the population-optimal UDM predictor generally depends on time: time controls how much the model should trust the observed context. We then show that this dependence can become negligible in finite-data settings relevant to language. When a corrupted training sequence remains much closer to its original clean sequence than to competing training sequences, the empirical-optimal predictor is nearly insensitive to time over most of the diffusion trajectory, where the guarantee weakens toward the high-noise endpoint. Empirically, trained language UDMs exhibit limited time sensitivity over most of the trajectory, while time-agnostic predictors remain competitive with, and often outperform, time-conditioned models across datasets and training objectives. These results challenge the use of explicit time conditioning in UDMs: although the population optimum depends on time, explicitly conditioning on it may often be unnecessary in practice.

Related

• From Truncation to Commitment: Persistent Context in Uniform Discrete Diffusion
• Plug-and-Play Guidance for Discrete Diffusion Models via Gradient-Informed Logit Correction
• On the Error-Correcting Effects of Stochasticity in Discrete Diffusion
• TAD: Temporal-Aware Trajectory Self-Distillation for Fast and Accurate Diffusion LLM

Source: arXiv cs.AI | 2026-09-28