D5P4: Partition Determinantal Point Process for Diversity in Parallel Discrete Diffusion Decoding

ArXi:2603.19146v1 Announce Type: new Discrete diffusion models are promising alternatives to autoregressive approaches for text generation, yet their decoding methods remain under-studied. Standard decoding methods for autoregressive models, such as beam search, do not directly apply to iterative denoising, and existing diffusion decoding techniques provide limited control over in-batch diversity. To bridge this gap, we