Precise: SDE-Consistent Stochastic Sampling for RL Post-Training of Flow-Matching Models
cs.LG, cs.AI, cs.CV
Submitted: 2026-05-22
Updated: 2026-11-08
Code: https://github.com/Tencent-Hunyuan/Precise
Terminology
Sources
- Training Diffusion Models with Reinforcement Learning
- HunyuanImage 3.0 Technical Report
- TreeGRPO: Tree-Advantage GRPO for Online RL Post-Training of Diffusion Models
- Seedance 1.0: Exploring the Boundaries of Video Generation Models
- HunyuanVideo: A Systematic Framework For Large Video Generative Models
- MixGRPO: Unlocking Flow-based GRPO Efficiency with Mixed ODE-SDE
- BranchGRPO: Stable and Efficient GRPO with Structured Branching in Diffusion Models
- Flow Matching for Generative Modeling
- Flow-GRPO: Training Flow Matching Models via Online RL
- Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow
- Latent Consistency Models: Synthesizing High-Resolution Images with Few-Step Inference
- Progressive Distillation for Fast Sampling of Diffusion Models
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Denoising Diffusion Implicit Models
- Score-Based Generative Modeling through Stochastic Differential Equations
- Wan: Open and Advanced Large-Scale Video Generative Models
- Coefficients-Preserving Sampling for Reinforcement Learning with Flow Matching
- GRPO-Guard: Mitigating Implicit Over-Optimization in Flow Matching via Regulated Clipping
- Unified Reward Model for Multimodal Understanding and Generation
- HunyuanVideo 1.5 Technical Report
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