Verified80% confidenceFactExact time
Flow matching is a class of methods for training continuous-time generative models that learn a velocity field to transport a prior distribution along ODE trajectories to the data distribution, and at inference time a numerical integrator is used to solve iteratively, with the number of integration steps being the NFE.
4
Sources
80%
Confidence
Long-term
Relevance
11/8/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedConditional flow matching (CFM) is a training objective under the flow matching framework that fits the instantaneous velocity of a path by constructing a continuous probability path from a source distribution to a target distribution (typically a straight-line interpolation path) and learning a time-dependent velocity field.80% similarUnverifiedCoupledFlow adopts a structure-constrained flow matching algorithm; unlike step-by-step iterative diffusion-style generation, flow matching achieves fast sampling by learning a velocity field, thereby enabling one-step prediction.79% similarVerifiedThe idea of flow matching is to learn a time-dependent vector field v_t(x) such that the ordinary differential equation driven by this vector field can transport a simple reference distribution to the target data distribution.79% similarVerifiedFlow matching models generate images by learning a continuous probability flow (ODE trajectory) from the noise distribution to the data distribution, and are generally more stable to train and require fewer inference steps.78% similarVerifiedFlow matching is a generative model training method that directly learns a deterministic ODE path from the noise distribution to the data distribution, defines a linear interpolation path xt=(1-t)x₀+tx₁, and uses a neural network to learn to predict the velocity vector at every point along that path.78% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1148309API
curl https://kongchang.com/api/v1/knowledge/claims/1148309MCP
get_claim(id=1148309)