Exploring the Design Space of Permutation Diffusion LLMs

Overview

arXiv:2610.20635v1 Announce Type: new Abstract: Permutation generation with language models requires both text-conditioned ranking and a global one-to-one assignment between candidates and output positions. We study this problem through three diffusion-LLM parameterizations: hard-assignment prediction over shuffled candidate-index sequences, soft-assignment prediction over doubly stochastic matrices, and direct ordering-score prediction followed by sorting. The three families place permutation c

SpinGraph analysis pending — check back after processing.