FewStepRL: Fine-tuning Few-step Diffusion Models with Reinforcement Learning
cs.LG, cs.CV
Submitted: 2026-11-05
Updated: 2026-11-05
Code: https://github.com/black-forest-labs/flux
Terminology
Sources
- Reinforce Adjoint Matching: Scaling RL Post-Training of Diffusion and Flow-Matching Models
- ShareGPT-4o-Image: Aligning Multimodal Models with GPT-4o-Level Image Generation
- Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scaling
- $R_\text{dm}$: Re-conceptualizing Distribution Matching as a Reward for Diffusion Distillation
- AnyFlow: Any-Step Video Diffusion Model with On-Policy Flow Map Distillation
- MeanFlowNFT: Bringing Forward-Process RL to Average-Velocity Generators
- GenEval 2: Addressing Benchmark Drift in Text-to-Image Evaluation
- Aligning Text-to-Image Models using Human Feedback
- Reinforcement Learning and Control as Probabilistic Inference: Tutorial and Review
- SDXL-Lightning: Progressive Adversarial Diffusion Distillation
- Continuous-Time Distribution Matching for Few-Step Diffusion Distillation
- Simplifying, Stabilizing and Scaling Continuous-Time Consistency Models
- LongCat-Video Technical Report
- GPT-4o System Card
- Proximal Policy Optimization Algorithms
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Seedream 4.0: Toward Next-generation Multimodal Image Generation
- Wan: Open and Advanced Large-Scale Video Generative Models
- Coefficients-Preserving Sampling for Reinforcement Learning with Flow Matching
- Exploring the Design Space of Reward Backpropagation for Flow Matching
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis