SIFR: Stable Signed Flow Reinforcement
cs.LG, cs.AI, cs.CV
Submitted: 2026-11-06
Updated: 2026-11-06
Code: https://github.com/christophschuhmann/improved-aesthetic-predictor
Terminology
Sources
- Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
- Reinforce Adjoint Matching: Scaling RL Post-Training of Diffusion and Flow-Matching Models
- Adjoint Matching: Fine-tuning Flow and Diffusion Generative Models with Memoryless Stochastic Optimal Control
- Online Reward-Weighted Fine-Tuning of Flow Matching with Wasserstein Regularization
- FlowAWR: Online Adaptive Flow Reinforcement via Advantage-Weighted Rectification
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- Classifier-Free Diffusion Guidance
- Knowledge Unlearning for Mitigating Privacy Risks in Language Models
- AdvantageFlow: Regularized Advantage-Weighted RL in Flow Models
- Reward Score Matching: Unifying Reward-based Fine-tuning for Flow and Diffusion Models
- Signed Rectified Flow: Negativity-Controlled Generation
- Flow-GRPO: Training Flow Matching Models via Online RL
- Improving Video Generation with Human Feedback
- Rectified Flow: A Marginal Preserving Approach to Optimal Transport
- TOFU: A Task of Fictitious Unlearning for LLMs
- Cosmos 3: Omnimodal World Models for Physical AI
- Advantage-Weighted Regression: Simple and Scalable Off-Policy Reinforcement Learning
- Aligning Text-to-Image Diffusion Models with Reward Backpropagation
- Proximal Policy Optimization Algorithms
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
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