Diffusion Mental Averages

TL;DR

This paper introduces Diffusion Mental Averages (DMA), a novel method for generating sharp, realistic concept averages within diffusion models by aligning denoising trajectories, extending to multimodal concepts with clustering and adaptation techniques.

Contribution

DMA is the first approach to produce consistent, realistic averages within diffusion models, capturing abstract concepts and serving as a visual summary and bias analysis tool.

Findings

DMA produces sharp, realistic concept averages.

The method extends to multimodal concepts using clustering and adaptation.

DMA offers insights into model biases and concept representations.

Abstract

Can a diffusion model produce its own "mental average" of a concept-one that is as sharp and realistic as a typical sample? We introduce Diffusion Mental Averages (DMA), a model-centric answer to this question. While prior methods aim to average image collections, they produce blurry results when applied to diffusion samples from the same prompt. These data-centric techniques operate outside the model, ignoring the generative process. In contrast, DMA averages within the diffusion model's semantic space, as discovered by recent studies. Since this space evolves across timesteps and lacks a direct decoder, we cast averaging as trajectory alignment: optimize multiple noise latents so their denoising trajectories progressively converge toward shared coarse-to-fine semantics, yielding a single sharp prototype. We extend our approach to multimodal concepts (e.g., dogs with many breeds) by clustering samples in semantically-rich spaces such as CLIP and applying Textual Inversion or LoRA to bridge CLIP clusters into diffusion space. This is, to our knowledge, the first approach that delivers consistent, realistic averages, even for abstract concepts, serving as a concrete visual summary and a lens into model biases and concept representation.