# Towards Better Alignment: Training Diffusion Models with Reinforcement Learning Against Sparse Rewards

arXiv: https://arxiv.org/abs/2503.11240 · HTML: https://mlsanity.com/paper/2503.11240/

## Key facts

- Cited by 32 papers in the map (data through 14 Oct 2026).
- Placed in Diffusion and flow policies for robotics, one of 14 directions in Diffusion models.
- Builds on 57 of its 72 references that are in the map.
- First posted on arXiv 14 Mar 2025.

## Placed in

- Diffusion models / [Diffusion and flow policies for robotics](https://mlsanity.com/diffusion/diffusion-and-flow-policies-for-robotics/)

## Builds on (most cited first, 30 of 57)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Diffusion Models Beat GANs on Image Synthesis](https://mlsanity.com/paper/2105.05233/) (2021)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding](https://mlsanity.com/paper/2205.11487/) (2022)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [Generative Adversarial Networks](https://mlsanity.com/paper/2203.00667/) (2022)
- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [Rethinking the Inception Architecture for Computer Vision](https://mlsanity.com/paper/1512.00567/) (2015)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [Zero-Shot Text-to-Image Generation](https://mlsanity.com/paper/2102.12092/) (2021)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Improved Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2102.09672/) (2021)
- [DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation](https://mlsanity.com/paper/2208.12242/) (2022)
- [DeBERTa: Decoding-enhanced BERT with Disentangled Attention](https://mlsanity.com/paper/2006.03654/) (2020)
- [Cascaded Diffusion Models for High Fidelity Image Generation](https://mlsanity.com/paper/2106.15282/) (2021)
- [Diffusion Models: A Comprehensive Survey of Methods and Applications](https://mlsanity.com/paper/2209.00796/) (2022)
- [Multi-Concept Customization of Text-to-Image Diffusion](https://mlsanity.com/paper/2212.04488/) (2022)
- [Improved Techniques for Training Score-Based Generative Models](https://mlsanity.com/paper/2006.09011/) (2020)
- [Diffusion-LM Improves Controllable Text Generation](https://mlsanity.com/paper/2205.14217/) (2022)
- [Training Diffusion Models with Reinforcement Learning](https://mlsanity.com/paper/2305.13301/) (2023)
- [Diffusion Model Alignment Using Direct Preference Optimization](https://mlsanity.com/paper/2311.12908/) (2023)
- [Data-Efficient Hierarchical Reinforcement Learning](https://mlsanity.com/paper/1805.08296/) (2018)
- [DPOK: Reinforcement Learning for Fine-tuning Text-to-Image Diffusion Models](https://mlsanity.com/paper/2305.16381/) (2023)

## Built on by (most cited first, 30 of 32)

- [X-Omni: Reinforcement Learning Makes Discrete Autoregressive Image Generative Models Great Again](https://mlsanity.com/paper/2507.22058/) (2025)
- [D-Fusion: Direct Preference Optimization for Aligning Diffusion Models with Visually Consistent Samples](https://mlsanity.com/paper/2505.22002/) (2025)
- [Integrating Reinforcement Learning with Visual Generative Models: Foundations and Advances](https://mlsanity.com/paper/2508.10316/) (2025)
- [Training-Free Adaptation of Diffusion Models via Doob's h-Transform](https://mlsanity.com/paper/2602.16198/) (2026)
- [Reinforcement Learning for Large Model: A Survey](https://mlsanity.com/paper/2508.08189/) (2025)
- [Alleviating Sparse Rewards by Modeling Step-Wise and Long-Term Sampling Effects in Flow-Based GRPO](https://mlsanity.com/paper/2602.06422/) (2026)
- [Self-Reflective Reinforcement Learning for Diffusion-based Image Reasoning Generation](https://mlsanity.com/paper/2505.22407/) (2025)
- [Mitigating Diffusion Model Hallucinations with Dynamic Guidance](https://mlsanity.com/paper/2510.05356/) (2025)
- [Diffusion Fine-Tuning via Reparameterized Policy Gradient of the Soft Q-Function](https://mlsanity.com/paper/2512.04559/) (2025)
- [Beyond the Dirac Delta: Mitigating Diversity Collapse in Reinforcement Fine-Tuning for Versatile Image Generation](https://mlsanity.com/paper/2601.12401/) (2026)
- [SCRIPT: Scalable Diffusion Policy with Multi-stage Training for Language-driven Physics-Based Humanoid Control](https://mlsanity.com/paper/2605.22894/) (2026)
- [Faster and Better Alignment for Flow Matching Models via Step-aware Advantages](https://mlsanity.com/paper/2602.01591/) (2026)
- [Rethinking Direct Preference Optimization in Diffusion Models](https://mlsanity.com/paper/2505.18736/) (2025)
- [Asynchronous Denoising Diffusion Models for Aligning Text-to-Image Generation](https://mlsanity.com/paper/2510.04504/) (2025)
- [DreamBoothDPO: Improving Personalized Generation using Direct Preference Optimization](https://mlsanity.com/paper/2505.20975/) (2025)
- [DP2O-SR: Direct Perceptual Preference Optimization for Real-World Image Super-Resolution](https://mlsanity.com/paper/2510.18851/) (2025)
- [StableSketcher: Enhancing Diffusion Model for Pixel-based Sketch Generation via Visual Question Answering Feedback](https://mlsanity.com/paper/2510.20093/) (2025)
- [Generative AI Meets 6G and Beyond: Diffusion Models for Semantic Communications](https://mlsanity.com/paper/2511.08416/) (2025)
- [ProxT2I: Efficient Reward-Guided Text-to-Image Generation via Proximal Diffusion](https://mlsanity.com/paper/2511.18742/) (2025)
- [HyperAlign: Hyperbolic Entailment Cones for Adaptive Text-to-Image Alignment Assessment](https://mlsanity.com/paper/2601.04614/) (2026)
- [OARS: Process-Aware Online Alignment for Generative Real-World Image Super-Resolution](https://mlsanity.com/paper/2603.12811/) (2026)
- [Offline Preference Optimization for Rectified Flow with Noise-Tracked Pairs](https://mlsanity.com/paper/2605.09433/) (2026)
- [Rhetorical Text-to-Image Generation via Two-layer Diffusion Policy Optimization](https://mlsanity.com/paper/2505.22792/) (2025)
- [HierKick: Hierarchical Reinforcement Learning for Vision-Guided Soccer Robot Control](https://mlsanity.com/paper/2603.00948/) (2026)
- [STARE: Step-wise Temporal Alignment and Red-teaming Engine for Multi-modal Toxicity Attack](https://mlsanity.com/paper/2605.00699/) (2026)
- [Fine-tuning Pocket-Aware Diffusion Models via Denoising Policy Optimization](https://mlsanity.com/paper/2605.17693/) (2026)
- [DivRL: Disentangled Self-Similarity Rewards for Diverse Subject-Driven Generation](https://mlsanity.com/paper/2606.23950/) (2026)
- [Selective Timestep Weighting and Advantage-Based Replay for Sample-Efficient Diffusion RLHF](https://mlsanity.com/paper/2607.07693/) (2026)
- [LeanGRPO: Eliminating Redundant Recomputation in Diffusion RL](https://mlsanity.com/paper/2609.03528/) (2026)
- [Compositional Reward Models for Conditional Medical Image Generation](https://mlsanity.com/paper/2609.05028/) (2026)

## Abstract (arXiv)

Diffusion models have achieved remarkable success in text-to-image generation. However, their practical applications are hindered by the misalignment between generated images and corresponding text prompts. To tackle this issue, reinforcement learning (RL) has been considered for diffusion model fine-tuning. Yet, RL's effectiveness is limited by the challenge of sparse reward, where feedback is only available at the end of the generation process. This makes it difficult to identify which actions during the denoising process contribute positively to the final generated image, potentially leading to ineffective or unnecessary denoising policies. To this end, this paper presents a novel RL-based framework that addresses the sparse reward problem when training diffusion models. Our framework, named B2-DiffuRL, employs two strategies: Backward progressive training and Branch-based sampling. For one thing, backward progressive training focuses initially on the final timesteps of denoising process and gradually extends the training interval to earlier timesteps, easing the learning difficulty from sparse rewards. For another, we perform branch-based sampling for each training interval. By comparing the samples within the same branch, we can identify how much the policies of the current training interval contribute to the final image, which helps to learn effective policies instead of unnecessary ones. B2-DiffuRL is compatible with existing optimization algorithms. Extensive experiments demonstrate the effectiveness of B2-DiffuRL in improving prompt-image alignment and maintaining diversity in generated images. The code for this work is available.