Flow-OPD: On-Policy Distillation for Flow Matching Models

• URL: http://arxiv.org/abs/2605.08063v3
• Date: Thu, 14 May 2026 12:16:32 GMT
• Title: Flow-OPD: On-Policy Distillation for Flow Matching Models
• Abstract summary: Flow-OPD is a post-training framework that integrates on-policy distillation into Flow Matching models.<n>It raises the GenEval score from 63 to 92 and the OCR accuracy from 59 to 94, yielding an overall improvement of roughly 10 points over vanilla GRPO.<n>These results establish Flow-OPD as a scalable alignment paradigm for building generalist text-to-image models.
• Score: 42.290892456464896
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Existing Flow Matching (FM) text-to-image models suffer from two critical bottlenecks under multi-task alignment: the reward sparsity induced by scalar-valued rewards, and the gradient interference arising from jointly optimizing heterogeneous objectives, which together give rise to a 'seesaw effect' of competing metrics and pervasive reward hacking. Inspired by the success of On-Policy Distillation (OPD) in the large language model community, we propose Flow-OPD, the first unified post-training framework that integrates on-policy distillation into Flow Matching models. Flow-OPD adopts a two-stage alignment strategy: it first cultivates domain-specialized teacher models via single-reward GRPO fine-tuning, allowing each expert to reach its performance ceiling in isolation; it then establishes a robust initial policy through a Flow-based Cold-Start scheme and seamlessly consolidates heterogeneous expertise into a single student via a three-step orchestration of on-policy sampling, task-routing labeling, and dense trajectory-level supervision. We further introduce Manifold Anchor Regularization (MAR), which leverages a task-agnostic teacher to provide full-data supervision that anchors generation to a high-quality manifold, effectively mitigating the aesthetic degradation commonly observed in purely RL-driven alignment. Built upon Stable Diffusion 3.5 Medium, Flow-OPD raises the GenEval score from 63 to 92 and the OCR accuracy from 59 to 94, yielding an overall improvement of roughly 10 points over vanilla GRPO, while preserving image fidelity and human-preference alignment and exhibiting an emergent 'teacher-surpassing' effect. These results establish Flow-OPD as a scalable alignment paradigm for building generalist text-to-image models. The codes and weights will be released in: https://github.com/CostaliyA/Flow-OPD .

Related papers

• Steering the Flow: Inverting Face Recognition Models via Gradient-Guided Flow Matching [17.11318422163162]
  Model Inversion Attacks aim to reconstruct identities from face recognition models, exposing critical security vulnerabilities.<n>We propose a novel two-stage white-box model inversion method that reformulates inversion as a trajectory-steering task.<n>Under an identity-disjoint cross-evaluation setting, SFMI achieves an ACC of 0.9248, an FID of 22.61, and an LPIPS of 0.3874 on the ArcFace target.
  arXiv  Detail & Related papers  (2026-08-17T16:47:09Z)
• Curvature-Adaptive Consistency Flow Matching: Autonomous Trajectory Optimization via Reinforcement Learning [8.422673935994451]
  consistency distillation has significantly accelerated the inference of diffusion models.<n>We propose Curvature-Adaptive Consistency Flow Matching to address the limitations of static sampling.<n>Our approach achieves new state-of-the-art results on large-scale models such as FLUX and SDXL.
  arXiv  Detail & Related papers  (2026-06-21T08:49:50Z)
• MoASE++: Mixture of Activation Sparsity Experts with Domain-Adaptive On-policy Distillation for Continual Test Time Adaptation [65.068801413044]
  Continual test-time adaptation adapts a source-pretrained model to non-stationary, unlabeled target streams.<n>We introduce MoASE, a mixture-of-experts that disentangles domain-agnostic structure from domain-specific texture.
  arXiv  Detail & Related papers  (2026-05-18T01:52:12Z)
• DiffusionOPD: A Unified Perspective of On-Policy Distillation in Diffusion Models [55.01951088768769]
  DiffusionOPD is a new multi-task training paradigm for diffusion models based on Online Policy Distillation (OPD)<n>We show that DiffusionOPD consistently surpasses both multi-reward RL and cascade RL baselines in training efficiency and final performance.
  arXiv  Detail & Related papers  (2026-05-14T16:49:09Z)
• Improved techniques for fine-tuning flow models via adjoint matching: a deterministic control pipeline [7.967038299436285]
  We propose a deterministic adjoint matching framework that formulates human preference alignment for flow-based generative models.<n>Building on this perspective, we introduce a truncated adjoint scheme that focuses on the terminal portion of the trajectory.<n>We further generalize the framework beyond standard KL-based regularization, allowing more flexible trade-offs between alignment strength and distributional preservation.
  arXiv  Detail & Related papers  (2026-05-07T17:12:47Z)
• Too Correct to Learn: Reinforcement Learning on Saturated Reasoning Data [55.84428098924793]
  We propose Constrained Uniform Top-K Sampling (CUTS), a parameter-free decoding strategy enforcing structure-preserving exploration.<n>We integrate this into Mixed-CUTS, a training framework synergizing exploitative and exploratory rollouts to amplify intra-group advantage variance.<n> Notably, Mixed-CUTS improves Pass@1 accuracy on the challenging AIME25 benchmark by up to 15.1% over standard GRPO.
  arXiv  Detail & Related papers  (2026-04-20T16:43:28Z)
• UniGRPO: Unified Policy Optimization for Reasoning-Driven Visual Generation [51.41441081823758]
  Unified models capable of interleaved generation have emerged as a promising paradigm.<n>We propose a unified reinforcement learning framework tailored for interleaved generation.<n>Our experiments demonstrate that this unified training recipe significantly enhances image generation quality through reasoning.
  arXiv  Detail & Related papers  (2026-03-24T17:59:17Z)
• Active Flow Matching [14.437387789022354]
  Active Flow Matching (AFM) reformulates variational objectives to operate on conditional endpoint distributions along the flow.<n>We derive forward and reverse Kullback-Leibler (KL) variants using self-normalised importance sampling.
  arXiv  Detail & Related papers  (2026-03-01T02:50:07Z)
• Euphonium: Steering Video Flow Matching via Process Reward Gradient Guided Stochastic Dynamics [49.242224984144904]
  We propose Euphonium, a novel framework that steers generation via process reward gradient guided dynamics.<n>Our key insight is to formulate the sampling process as a theoretically principled algorithm that explicitly incorporates the gradient of a Process Reward Model.<n>We derive a distillation objective that internalizes the guidance signal into the flow network, eliminating inference-time dependency on the reward model.
  arXiv  Detail & Related papers  (2026-02-04T08:59:57Z)
• Dual-Stream Diffusion for World-Model Augmented Vision-Language-Action Model [62.889356203346985]
  We propose DUal-STream diffusion (DUST), a world-model augmented VLA framework that handles the modality conflict.<n>DUST achieves up to 6% gains over a standard VLA baseline and implicit world-modeling methods.<n>On real-world tasks with the Franka Research 3, DUST outperforms baselines in success rate by 13%.
  arXiv  Detail & Related papers  (2025-10-31T16:32:12Z)
• Fine-tuning Flow Matching Generative Models with Intermediate Feedback [23.447961856739422]
  Flow-based generative models have shown remarkable success in text-to-image generation.<n>Existing approaches solely learn from outcome rewards, struggling with the credit assignment problem.<n>We present AC-Flow, a robust actor-critic framework that addresses these challenges through three key innovations.
  arXiv  Detail & Related papers  (2025-10-20T20:08:03Z)
• Adaptive Divergence Regularized Policy Optimization for Fine-tuning Generative Models [31.470613363668672]
  Adaptive Divergence Regularized Policy Optimization automatically adjusts regularization strength based on advantage estimates.<n>Our implementation with Wasserstein-2 regularization for flow matching generative models achieves remarkable results on text-to-image generation.<n> ADRPO generalizes to KL-regularized fine-tuning of both text-only LLMs and multi-modal reasoning models.
  arXiv  Detail & Related papers  (2025-10-20T19:46:02Z)
• Improving Progressive Generation with Decomposable Flow Matching [50.63174319509629]
  Decomposable Flow Matching (DFM) is a simple and effective framework for the progressive generation of visual media.<n>On Imagenet-1k 512px, DFM achieves 35.2% improvements in FDD scores over the base architecture and 26.4% over the best-performing baseline.
  arXiv  Detail & Related papers  (2025-06-24T17:58:02Z)
• Flow-GRPO: Training Flow Matching Models via Online RL [80.62659379624867]
  We propose Flow-GRPO, the first method to integrate online policy reinforcement learning into flow matching models.<n>Our approach uses two key strategies: (1) an ODE-to-SDE conversion that transforms a deterministic Ordinary Differential Equation into an equivalent Differential Equation (SDE) that matches the original model's marginal distribution at all timesteps; and (2) a Denoising Reduction strategy that reduces training denoising steps while retaining the original number of inference steps.
  arXiv  Detail & Related papers  (2025-05-08T17:58:45Z)
• Online Reward-Weighted Fine-Tuning of Flow Matching with Wasserstein Regularization [14.320131946691268]
  We propose an easy-to-use and theoretically sound fine-tuning method for flow-based generative models.<n>By introducing an online rewardweighting mechanism, our approach guides the model to prioritize high-reward regions in the data manifold.<n>Our method achieves optimal policy convergence while allowing controllable trade-offs between reward and diversity.
  arXiv  Detail & Related papers  (2025-02-09T22:45:15Z)

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.