Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Diffusion Domain Expansion: Learning to Coordinate Pre-trained Diffusion Models

About

In this paper, we propose Diffusion Domain Expansion (DDE), a method that efficiently extends pre-trained diffusion models to generate larger objects and handle more complex conditioning beyond their original capabilities. Our method employs a compact trainable network designed to coordinate the denoised outputs of pre-trained diffusion models. We demonstrate that the coordinator can be universally simple while being capable of generalizing to domains larger than those observed during its training time. We evaluate DDE on long audio track generation and conditional image generation, demonstrating its applicability across domains. DDE outperforms other approaches to coordinated generation with diffusion models in qualitative and quantitative evaluations.

Egor Lifar, Semyon Savkin, Timur Garipov, Shangyuan Tong, Tommi Jaakkola• 2026

Related benchmarks

TaskDatasetResultRank
Conditional Image GenerationCLEVR
Accuracy96.5
25
Long Music Track GenerationSlakh2100 4l length (test)
FAD2.112
5
Long Music Track GenerationSlakh2100 10l length (test)
FAD2.142
5
Satellite image generationGoogle Maps (test)
FID27.373
4
Showing 4 of 4 rows

Other info

Follow for update