Skip to content
AI News HubLIVE
Original source2 min read

Trajectory-Regularized Distribution Distillation for One-Step Generation

Summary

The paper proposes Trajectory-Regularized Distribution Distillation (TRD), which combines distribution-level and trajectory-level supervision via a momentum-mixture policy predicted in a single student forward pass. On SANA-1.6B and Stable Diffusion 3.5, TRD outperforms consistency and distribution matching baselines in generation quality and training efficiency for one-step and few-step generation, with a trade-off in sample diversity.

SourcearXiv Computer VisionAuthor: An Zhao, Yixiang Zhou, Zejian Li, Yi Guo, Pei Chen, Lingyun Sun
Trajectory-Regularized Distribution Distillation for One-Step Generation
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions
Read article

[Submitted on 8 Nov 2026]

Title:Trajectory-Regularized Distribution Distillation for One-Step Generation

View a PDF of the paper titled Trajectory-Regularized Distribution Distillation for One-Step Generation, by An Zhao and 5 other authors

View PDF

Abstract:Distilling flow matching models for one-step generation generally follows two paradigms. Trajectory alignment supervises the student with the teacher's probability flow ODE, but its performance degrades under very small sampling budgets. Distribution matching directly optimizes the generated distribution, but estimating its reverse Kullback-Leibler gradient typically requires a separate network to track the evolving student score. We propose trajectory-regularized distribution distillation (TRD), which combines distribution-level and trajectory-level supervision through a momentum-mixture policy predicted in a single student forward pass. This policy serves simultaneously as a one-step generator, an estimator of the student's marginal velocity that removes the need for a second trainable network in distribution matching, and a closed-form time-dependent velocity field supervised along the teacher ODE trajectory. Experiments on SANA-1.6B and Stable Diffusion 3.5 show that TRD outperforms representative consistency and distribution matching baselines in generation quality and training efficiency for both one-step and few-step generation, with a trade-off in sample diversity.

Subjects:

Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)

Cite as: arXiv:2611.12978 [cs.CV]

(or arXiv:2611.12978v1 [cs.CV] for this version)

https://doi.org/10.48550/arXiv.2611.12978

arXiv-issued DOI via DataCite

Submission history

From: An Zhao [view email] [v1] Sun, 8 Nov 2026 06:05:40 UTC (16,498 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled Trajectory-Regularized Distribution Distillation for One-Step Generation, by An Zhao and 5 other authors

View PDF

TeX Source

view license

Additional Features

Audio Summary

Current browse context:

cs.CV

new | recent | 2026-11

Change to browse by:

cs cs.AI

References & Citations

SciX

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?)

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?)

Key points and analysis

Article intelligence

EngineersAdvanced

Key points

  • TRD unifies distribution matching and trajectory alignment using a single momentum-mixture policy that acts as a one-step generator, marginal velocity estimator, and closed-form time-dependent velocity field.
  • The method eliminates the need for a second trainable network to track the student score in distribution matching, improving training efficiency.
  • Experiments on SANA-1.6B and Stable Diffusion 3.5 show TRD surpasses representative consistency and distribution matching baselines in quality and efficiency for one-step and few-step generation.
  • TRD achieves better generation quality and training efficiency at the cost of reduced sample diversity.

Highlights and analysis are generated automatically and may contain errors. Check the original source.