Review
Oct 2026
When Diffusion Recommenders Stop Sampling: Preserving Generative Diversity
A user history can admit multiple plausible next interactions, yet conventional sequential recommenders produce a single deterministic ranking. While diffusion models offer a natural way to sample alternative continuations, we show that existing target-item diffusion recommenders often suffer from severe sampling colla...
Louis Castres Saint Martin, Alain Rakotomamonjy, P. Gallinari
· 0 citations