Primal-Dual Guided Decoding for Constrained Discrete Diffusion

arXiv:2605.09749v1 Announce Type: new Abstract: Discrete diffusion models generate structured sequences by progressively unmasking tokens, but enforcing global property constraints during generation remains an open challenge. We propose primal-dual guided decoding, an inference-time method that formulates constrained generation as a KL-regularised optimisation problem and solves it online via adaptive Lagrangian multipliers. At each denoising step, the method modifies token logits through an add