Follow the Mean: Reference-Guided Flow Matching

TL;DR

This paper introduces a novel reference-guided flow matching method for controllable generation, enabling model steering through reference sets without fine-tuning or auxiliary networks.

Contribution

It presents a simple, training-free control principle using reference means and a semi-parametric approach with residual refinement, allowing flexible, data-driven model adaptation.

Findings

Control of color, identity, style, and structure achieved with frozen models.

Reference-Mean Guidance operates without additional training, using a closed-form correction.

Semi-Parametric Guidance matches state-of-the-art quality while enabling reference set swapping.

Abstract

Existing approaches to controllable generation typically rely on fine-tuning, auxiliary networks, or test-time search. We show that flow matching admits a different control interface: adaptation through examples. For deterministic interpolants, the velocity field is solely governed by a conditional endpoint mean; shifting this mean shifts the flow itself. This yields a simple principle for controllable generation: steer a pretrained model by changing the reference set it follows. We instantiate this idea in two forms. Reference-Mean Guidance is training-free: it computes a closed-form endpoint-mean correction from a reference bank and applies it to a frozen FLUX.2-klein (4B) model, enabling control of color, identity, style, and structure while keeping the prompt, seed, and weights fixed. Semi-Parametric Guidance amortizes the same idea through an explicit mean anchor and learned residual refiner, matching unconditional DiT-B/4 quality on AFHQv2 while allowing the reference set to be swapped at inference time. These results point to a broader direction: generative models that adapt through data, not parameter updates.