FlowConsist: Make Your Flow Consistent with Real Trajectory

TL;DR

FlowConsist introduces a training framework for fast flow models that enforces trajectory consistency, significantly improving generation quality and speed by addressing trajectory drift and error accumulation.

Contribution

The paper proposes a novel training method that replaces conditional velocities with model-predicted marginal velocities and introduces trajectory rectification to enhance flow consistency.

Findings

Achieves state-of-the-art FID of 1.52 on ImageNet 256x256 with 1 sampling step.

Addresses trajectory drift and error accumulation in fast flow models.

Demonstrates improved sample quality and consistency over previous methods.

Abstract

Fast flow models accelerate the iterative sampling process by learning to directly predict ODE path integrals, enabling one-step or few-step generation. However, we argue that current fast-flow training paradigms suffer from two fundamental issues. First, conditional velocities constructed from randomly paired noise-data samples introduce systematic trajectory drift, preventing models from following a consistent ODE path. Second, the model's approximation errors accumulate over time steps, leading to severe deviations across long time intervals. To address these issues, we propose FlowConsist, a training framework designed to enforce trajectory consistency in fast flows. We propose a principled alternative that replaces conditional velocities with the marginal velocities predicted by the model itself, aligning optimization with the true trajectory. To further address error accumulation over time steps, we introduce a trajectory rectification strategy that aligns the marginal distributions of generated and real samples at every time step along the trajectory. Our method establishes a new state-of-the-art on ImageNet 256$\times$256, achieving an FID of 1.52 with only 1 sampling step.