pith. sign in

B.3 INTUITION BEHIND THEFLOWGRPO OBJECTIVE We provide some insight into reverse-process diffusion RL by inspecting the FlowGRPO objective in a sampler-agnostic manner

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

fields

cs.LG 1

years

2025 1

verdicts

UNVERDICTED 1

representative citing papers

DiffusionNFT: Online Diffusion Reinforcement with Forward Process

cs.LG · 2025-09-19 · unverdicted · novelty 7.0

DiffusionNFT performs online RL for diffusion models on the forward process via flow matching and positive-negative contrasts, delivering up to 25x efficiency gains and rapid benchmark improvements over prior reverse-process methods.

citing papers explorer

Showing 1 of 1 citing paper.

  • DiffusionNFT: Online Diffusion Reinforcement with Forward Process cs.LG · 2025-09-19 · unverdicted · none · ref 30

    DiffusionNFT performs online RL for diffusion models on the forward process via flow matching and positive-negative contrasts, delivering up to 25x efficiency gains and rapid benchmark improvements over prior reverse-process methods.