D2PO: Optimizing Diffusion Samplers via Dynamic Preference
D2PO (Dynamic Direct Preference Optimization) is a framework for optimizing diffusion sampling policies, including timestep schedules and classifier-free guidance weights. It addresses the limitation of student-teacher regression where low-NFE student samplers sacrifice high-frequency texture fidelity. By reformulating optimization as preference alignment using energy-based models and dynamic preferences, it achieves better perceptual quality. Experiments show superiority over regression schedulers under low-NFE constraints.
-->
[Submitted on 7 Jul 2026]
Title:D2PO: Optimizing Diffusion Samplers via Dynamic Preference
View a PDF of the paper titled D2PO: Optimizing Diffusion Samplers via Dynamic Preference, by Jinkyu Kim and 2 other authors
View PDF HTML (experimental)
Abstract:We propose D2PO (Dynamic Direct Preference Optimization), a principled framework for optimizing diffusion sampling policies with respect to timestep schedules and classifier-free guidance (CFG) weights. Our work is motivated by a fundamental limitation of existing student-teacher regression frameworks; low-NFE student samplers are trained to mimic high-NFEteachers, often sacrificing high-frequency texture fidelity while preserving coarse global structures, thereby misaligning the sampler with perceptual quality. D2PO addresses this challenge by reformulating sampler optimization as a preference-based alignment problem, leveraging the Direct Preference Optimization (DPO) framework. To make DPO applicable to diffusion samplers, we model the sampling policy as an energy-based model (EBM), transforming preference comparisons into tractable energy differences. We further introduce a novel energy formulation derived directly from the pretrained score network, enabling preference evaluation in perturbed spaces that jointly capture structural consistency and fine-grained details. Moreover, we introduce dynamic preferences, where the preferred samples used for alignment progressively improve as the sampling policies are learned. This self-improving mechanism replaces rigid static teacher supervision with an iterative, preference-guided refinement process, providing progressively stronger alignment signals. Extensive experiments demonstrate that D2PO aligns diffusion samplers with perceptual quality more faithfully, unlocking the full potential of high-quality teachers and consistently outperforming conventional regression-based schedulers under low-NFE constraints.
Comments: Accepted to ECCV 2026
Subjects:
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Cite as: arXiv:2607.06609 [cs.LG]
(or arXiv:2607.06609v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2607.06609
arXiv-issued DOI via DataCite
Submission history
From: Jinkyu Kim [view email] [v1] Tue, 7 Jul 2026 06:05:42 UTC (12,739 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled D2PO: Optimizing Diffusion Samplers via Dynamic Preference, by Jinkyu Kim and 2 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.LG
new | recent | 2026-07
Change to browse by:
cs cs.AI
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
Loading...
Data provided by:
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Code, Data and Media Associated with this Article
alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos
Demos
Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Related Papers
Recommenders and Search Tools
Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
IArxiv recommender toggle
IArxiv Recommender (What is IArxiv?)
Author
Venue
Institution
Topic
About arXivLabs
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)