Reward Score Matching: Unifying Reward-based Fine-tuning for Flow and Diffusion Models

• URL: http://arxiv.org/abs/2604.17415v1
• Date: Sun, 19 Apr 2026 12:47:52 GMT
• Title: Reward Score Matching: Unifying Reward-based Fine-tuning for Flow and Diffusion Models
• Abstract summary: Reward-based fine-tuning aims to steer a pretrained diffusion or flow-based generative model toward higher-reward samples.<n>We show that many can be written under a common framework, which we call reward score matching (RSM)
• Score: 54.597200388738656
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Reward-based fine-tuning aims to steer a pretrained diffusion or flow-based generative model toward higher-reward samples while remaining close to the pretrained model. Although existing methods are motivated by different perspectives such as Soft RL, GFlowNets, etc., we show that many can be written under a common framework, which we call reward score matching (RSM). Under this view, alignment becomes score matching toward a reward-guided target, and the main differences across methods reduce to the construction of the value-guidance estimator and the effective optimization strength across timesteps. This unification clarifies the bias--variance--compute tradeoffs of existing designs and distinguishes core optimization components from auxiliary mechanisms that add complexity without clear benefit. Guided by this perspective, we develop simpler redesigns that improve alignment effectiveness and compute efficiency across representative settings with differentiable and black-box rewards. Overall, RSM turns a seemingly fragmented collection of reward-based fine-tuning methods into a smaller, more interpretable, and more actionable design space.

Related papers

• Scaling Reinforcement Learning for Diffusion Models via Velocity Matching [66.34541945134916]
  RewardBench fine-tuning is an important tool for adapting diffusion models to human preferences and task-specific objectives.<n>Unlike autoregressive models, diffusion models do not provide tractable likelihoods for generated samples.<n>We propose reward-based velocity matching (RVM), a simple trajectory-free update that acts directly on the velocity field.
  arXiv  Detail & Related papers  (2026-08-24T17:13:43Z)
• Optimizing Visual Generative Models via Distribution-wise Rewards [57.46109819595665]
  We present a novel framework that finetunes generative models using distribution-wise rewards, ensuring better alignment with real-world data distributions.<n>Our approach significantly improves FID-50K across various base models, from 5.77 for SiT and from 3.74 to 3.52 for EDM2.
  arXiv  Detail & Related papers  (2026-07-02T15:08:56Z)
• Alternating Reinforcement Learning with Contextual Rubric Rewards [2.1172256884504588]
  Reinforcement Learning with Rewards (RLRR) is a framework that extends conventional reinforcement learning from human feedback.<n>Existing approaches in RLRR are limited to linearly compressing vector rewards into a scalar reward with a fixed weightings.<n>This work proposes Alternating Reinforcement Learning with Rewards (ARL-RR) to overcome the limitations of reward aggregation.
  arXiv  Detail & Related papers  (2026-03-04T04:18:39Z)
• Taming Preference Mode Collapse via Directional Decoupling Alignment in Diffusion Reinforcement Learning [27.33241821967005]
  We propose a novel framework that mitigates Preference Mode Collapse (PMC)<n>D$2$-Align achieves superior alignment with human preference.
  arXiv  Detail & Related papers  (2025-12-30T11:17:52Z)
• APLOT: Robust Reward Modeling via Adaptive Preference Learning with Optimal Transport [37.21695864040979]
  The reward model (RM) plays a crucial role in aligning Large Language Models (LLMs) with human preferences through Reinforcement Learning.<n>This paper introduces an effective enhancement to BT-based RMs through an adaptive margin mechanism.
  arXiv  Detail & Related papers  (2025-10-13T03:13:28Z)
• Divergence Minimization Preference Optimization for Diffusion Model Alignment [66.31417479052774]
  Divergence Minimization Preference Optimization (DMPO) is a principled method for aligning diffusion models by minimizing reverse KL divergence.<n>DMPO can consistently outperform or match existing techniques across different base models and test sets.
  arXiv  Detail & Related papers  (2025-07-10T07:57:30Z)
• Interpretable Reward Modeling with Active Concept Bottlenecks [54.00085739303773]
  We introduce Concept Bottleneck Reward Models (CB-RM), a reward modeling framework that enables interpretable preference learning.<n>Unlike standard RLHF methods that rely on opaque reward functions, CB-RM decomposes reward prediction into human-interpretable concepts.<n>We formalize an active learning strategy that dynamically acquires the most informative concept labels.
  arXiv  Detail & Related papers  (2025-07-07T06:26:04Z)
• Self-Boost via Optimal Retraining: An Analysis via Approximate Message Passing [58.52119063742121]
  Retraining a model using its own predictions together with the original, potentially noisy labels is a well-known strategy for improving the model performance.<n>This paper addresses the question of how to optimally combine the model's predictions and the provided labels.<n>Our main contribution is the derivation of the Bayes optimal aggregator function to combine the current model's predictions and the given labels.
  arXiv  Detail & Related papers  (2025-05-21T07:16:44Z)
• Self Distillation via Iterative Constructive Perturbations [0.2748831616311481]
  We propose a novel framework that uses a cyclic optimization strategy to concurrently optimize the model and its input data for better training.<n>By alternately altering the model's parameters to the data and the data to the model, our method effectively addresses the gap between fitting and generalization.
  arXiv  Detail & Related papers  (2025-05-20T13:15:27Z)
• Directly Fine-Tuning Diffusion Models on Differentiable Rewards [22.1685707267062]
  We present Direct Reward Fine-Tuning (DRaFT), a method for fine-tuning diffusion models to maximize differentiable reward functions. We first show that it is possible to backpropagate the reward function gradient through the full sampling procedure, and that doing so achieves strong performance on a variety of rewards. We show that our methods work well for a variety of reward functions and can be used to substantially improve the aesthetic quality of images generated by Stable Diffusion 1.4.
  arXiv  Detail & Related papers  (2023-09-29T17:01:02Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.