The Latent Color Subspace: Emergent Order in High-Dimensional Chaos

TL;DR

This paper uncovers a structured latent color subspace in high-dimensional autoencoder models, enabling explicit and training-free control over image color attributes in text-to-image generation.

Contribution

It introduces the Latent Color Subspace (LCS) interpretation, revealing a structured color encoding in the latent space and providing a novel, training-free color control method.

Findings

The LCS reflects Hue, Saturation, and Lightness in the latent space.

LCS can predict and control image color explicitly.

The method is training-free and based on closed-form latent-space manipulation.

Abstract

Text-to-image generation models have advanced rapidly, yet achieving fine-grained control over generated images remains difficult, largely due to limited understanding of how semantic information is encoded. We develop an interpretation of the color representation in the Variational Autoencoder latent space of FLUX.1 [Dev], revealing a structure reflecting Hue, Saturation, and Lightness. We verify our Latent Color Subspace (LCS) interpretation by demonstrating that it can both predict and explicitly control color, introducing a fully training-free method in FLUX based solely on closed-form latent-space manipulation. Code is available at https://github.com/ExplainableML/LCS.