# D-OPSD: On-Policy Self-Distillation for Continuously Tuning Step-Distilled Diffusion Models

arXiv: https://arxiv.org/abs/2605.05204 · HTML: https://mlsanity.com/paper/2605.05204/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models.
- Builds on 101 of its 118 references that are in the map.
- First posted on arXiv 6 May 2026.

## Placed in

- Diffusion models / [Diffusion foundations and sampling](https://mlsanity.com/diffusion/diffusion-foundations-and-sampling/)

## Builds on (most cited first, 30 of 101)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [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)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Score-Based Generative Modeling through Stochastic Differential Equations](https://mlsanity.com/paper/2011.13456/) (2020)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Diffusion Models Beat GANs on Image Synthesis](https://mlsanity.com/paper/2105.05233/) (2021)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (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)
- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [Scaling Rectified Flow Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.03206/) (2024)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow](https://mlsanity.com/paper/2209.03003/) (2022)
- [DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation](https://mlsanity.com/paper/2208.12242/) (2022)
- [Sequence to Sequence Learning with Neural Networks](https://mlsanity.com/paper/1409.3215/) (2014)
- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)

## Built on by (most cited first, 22 of 22)

- [DanceOPD: On-Policy Generative Field Distillation](https://mlsanity.com/paper/2606.27377/) (2026)
- [OPSD-V: On-Policy Self-Distillation for Post-Training Few-Step Autoregressive Video Generators](https://mlsanity.com/paper/2607.08766/) (2026)
- [OmniOPSD: Rationale-Privileged On-Policy Self-Distillation for Affective Computing](https://mlsanity.com/paper/2606.15920/) (2026)
- [An Empirical Study of Training Pixel-Space Text-to-Image Diffusion Models](https://mlsanity.com/paper/2608.16887/) (2026)
- [Rethinking Classifier-Free Guidance in On-Policy Diffusion Distillation](https://mlsanity.com/paper/2607.24731/) (2026)
- [CollectionLoRA: Collecting 50 Effects in 1 LoRA via Multi-Teacher On-Policy Distillation](https://mlsanity.com/paper/2605.25378/) (2026)
- [STEP-OPD: Rethinking Output Targets and Internal Dynamics in On-Policy Distillation for Diffusion Models](https://mlsanity.com/paper/2608.04887/) (2026)
- [Latent Reward Registers for Diffusion Preference Alignment](https://mlsanity.com/paper/2608.03929/) (2026)
- [HPSD: Hybrid-Policy Self-Distillation for Text-Image-to-Video Diffusion Models](https://mlsanity.com/paper/2608.13205/) (2026)
- [Self-OPD: On-Policy Distillation for Flow Matching Models without Teacher](https://mlsanity.com/paper/2608.26872/) (2026)
- [On-Policy Self-Distillation for Multi-Turn Image Editing](https://mlsanity.com/paper/2609.35611/) (2026)
- [From SRA to Self-Flow: Data Augmentation or Self-Supervision?](https://mlsanity.com/paper/2607.02508/) (2026)
- [ABOPD: Antibody CDR Design via On-Policy Distillation](https://mlsanity.com/paper/2607.18835/) (2026)
- [SelfLift: Accelerating Few-Step Diffusion via Self-Recovering Resolution Transition](https://mlsanity.com/paper/2609.02036/) (2026)
- [PE-OPSD: Internalizing Prompt Enhancement into Flow-matching Models via On-Policy Self-Distillation](https://mlsanity.com/paper/2609.36638/) (2026)
- [LIFT: Layout-In-Future Video Generation under Large Viewpoint Change via On-Policy Self-Distillation](https://mlsanity.com/paper/2609.38146/) (2026)
- [UniEvo-VL: An On-policy Self-Distillation Training Recipe for Multimodal Model Self-improvement](https://mlsanity.com/paper/2609.38721/) (2026)
- [Salvation Lies Within: Eliciting Inherent Style Transfer in Step-Distilled Diffusion Models](https://mlsanity.com/paper/2610.05066/) (2026)
- [Internalizing Agent Experience into Diffusion Model Weights via On-Policy Context Distillation](https://mlsanity.com/paper/2610.07250/) (2026)
- [ForcingBooster: Adaptive Patch-wise Denoising for Fast Autoregressive Video Diffusion](https://mlsanity.com/paper/2610.12625/) (2026)
- [A Lightweight Velocity Correction Framework for Improving Instruction Following in Few-Step Text-to-Image Generation](https://mlsanity.com/paper/2610.12899/) (2026)
- [InstantFusion: A Shared Latent Interface for Heterogeneous Image Generators](https://mlsanity.com/paper/2610.13944/) (2026)

## Abstract (arXiv)

The landscape of high-performance image generation models is currently shifting from the inefficient multi-step ones to the efficient few-step counterparts (e.g, Z-Image-Turbo and FLUX.2-klein). However, these models present significant challenges for direct continuous supervised fine-tuning. For example, applying the commonly used fine-tuning technique would compromise their inherent few-step inference capability. To address this, we propose D-OPSD, a novel training paradigm for step-distilled diffusion models that enables on-policy learning during supervised fine-tuning. We first find that the modern diffusion models, where the LLM/VLM serves as the encoder, can inherit its encoder's in-context capabilities. This enables us to formulate the training as an on-policy self-distillation process. Specifically, during training, we make the model act as both the teacher and the student with different contexts, where the student is conditioned only on the text feature, while the teacher is conditioned on the multimodal feature of both the text prompt and the target image. Training minimizes the two predicted distributions over the student's own roll-outs. By optimizing on the model's own trajectory and under its own supervision, D-OPSD enables the model to learn new concepts, styles, etc., without sacrificing the original few-step capacity.