AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Stepwise Credit Assignment for GRPO on Flow-Matching Models

Yash Savani, Branislav Kveton, Yuchen Liu, Yilin Wang, Jing Shi, Subhojyoti Mukherjee, Nikos Vlassis, Krishna Kumar Singh

CVPR 2026 · Image generation › Fast diffusion sampling · approach: Guidance and alignment for flow matching · 9 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Replace uniform final-image advantage with stepwise gain-based advantagesFlow-GRPO · GenEval (cfg=4.5)Overall accuracy: 0.68 -> 0.71Improved
Replace uniform final-image advantage with stepwise gain-based advantagesFlow-GRPO · GenEval (cfg=4.5)Counting accuracy: 0.64 -> 0.70Improved
Replace uniform final-image advantage with stepwise gain-based advantagesFlow-GRPO · GenEval (cfg=1.0)Overall accuracy: 0.60 -> 0.60No clear effect
Replace uniform final-image advantage with stepwise gain-based advantagesFlow-GRPO · Reward per training step (4 reward settings)RewardImproved
Add stepwise credit assignment on top of DDIM-inspired SDE (ablation)Flow-GRPO with DDIM SDE · Reward per training stepRewardImproved
Use DDIM-inspired SDE instead of Flow-GRPO SDE (ablation)Flow-GRPO and Stepwise-Flow-GRPO · Reward per training stepRewardImproved
Use 5 ODE substeps for intermediate reward estimates (vs 1 to 3) (ablation)Stepwise-Flow-GRPO · Reward estimate quality (appendix)Training signalImproved
Use more than 5 ODE substeps (6 to 10) (ablation)Stepwise-Flow-GRPO · Reward estimate quality (appendix)Training signalNo clear effect
Use joint normalization of gains instead of per-step normalization (ablation)Stepwise-Flow-GRPO · Convergence (appendix)Convergence speedImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy