Diffusion-MPC in Discrete Domains: Feasibility Constraints, Horizon Effects, and Critic Alignment: Case study with Tetris

Researchers studied diffusion-based model predictive control in discrete domains using Tetris, finding that feasibility constraints are necessary and shorter planning horizons outperform longer ones. The study reveals structural challenges with discrete diffusion planners, particularly misalignment issues with DQN critics that produce high decision regret.