Skip to content
AI News HubLIVE
Original source2 min read

A Lagrangian View of Flow Matching

Summary

arXiv paper 2609.00198 by Peyman Milanfar re-derives Flow Matching from a bottom-up, particle-centric Lagrangian perspective instead of the standard Eulerian optimal-transport route, motivating a target-identity conservation condition whose governing quasi-linear advection PDE yields the straight-line trajectories of Flow Matching and isolates the denoiser Jacobian as the source of trajectory curvature.

SourcearXiv Computer VisionAuthor: Peyman Milanfar
A Lagrangian View of Flow Matching
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 31 Aug 2026]

Title:A Lagrangian View of Flow Matching

View a PDF of the paper titled A Lagrangian View of Flow Matching, by Peyman Milanfar

View PDF HTML (experimental)

Abstract:Modern explicit-time generative models, such as Flow Matching [Lipman et al., 2023] and Rectified Flow [Liu et al., 2023], are typically derived top-down via Optimal Transport and the continuity equation. This standard Eulerian approach focuses on the macroscopic transport of probability mass. In this paper, we present an alternative, bottom-up mechanical derivation grounded in a Lagrangian (particle-centric) perspective. By analyzing the local Taylor expansion of a continuous denoiser, we motivate a strict invariance condition required for optimal, singlestep generation: the conservation of target identity. Enforcing this condition yields a governing quasi-linear advection Partial Differential Equation (PDE). We demonstrate that solving this PDE via the Method of Characteristics analytically yields the straight-line trajectories of Flow Matching. This geometric perspective isolates the Jacobian of the denoiser as the primary source of trajectory curvature, providing a direct mathematical explanation for why straight-line flows enable massive step sizes, and why empirical models require distillation to flatten intersecting characteristics.

Subjects:

Computer Vision and Pattern Recognition (cs.CV); Fluid Dynamics (physics.flu-dyn)

Cite as: arXiv:2609.00198 [cs.CV]

(or arXiv:2609.00198v1 [cs.CV] for this version)

https://doi.org/10.48550/arXiv.2609.00198

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Peyman Milanfar [view email] [v1] Mon, 31 Aug 2026 18:13:33 UTC (192 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled A Lagrangian View of Flow Matching, by Peyman Milanfar

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.CV

new | recent | 2026-09

Change to browse by:

cs physics physics.flu-dyn

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?)

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

  • Replaces the top-down Eulerian optimal-transport derivation with a bottom-up Lagrangian, particle-centric mechanical derivation of Flow Matching.
  • Motivates a strict target-identity conservation condition for optimal single-step generation, which yields a governing quasi-linear advection PDE.
  • Solving that PDE via the Method of Characteristics analytically reproduces Flow Matching's straight-line trajectories.
  • Identifies the denoiser Jacobian as the primary source of trajectory curvature, explaining why straight-line flows enable large step sizes and why empirical models need distillation.

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