GimmBO: Interactive Generative Image Model Merging via Bayesian Optimization

TL;DR

GimmBO introduces an interactive Bayesian optimization framework for merging image adapters, enabling efficient exploration of the complex merging space and improving image generation quality through user-guided optimization.

Contribution

It presents a novel two-stage Bayesian optimization method tailored for adapter merging in diffusion models, enhancing efficiency and user experience.

Findings

Improved convergence and success rates over baseline methods

Effective handling of high-dimensional adapter merging space

Demonstrated flexibility through various extensions

Abstract

Fine-tuning-based adaptation is widely used to customize diffusion-based image generation, leading to large collections of community-created adapters that capture diverse subjects and styles. Adapters derived from the same base model can be merged with weights, enabling the synthesis of new visual results within a vast and continuous design space. To explore this space, current workflows rely on manual slider-based tuning, an approach that scales poorly and makes weight selection difficult, even when the candidate set is limited to 20-30 adapters. We propose GimmBO to support interactive exploration of adapter merging for image generation through Preferential Bayesian Optimization (PBO). Motivated by observations from real-world usage, including sparsity and constrained weight ranges, we introduce a two-stage BO backend that improves sampling efficiency and convergence in high-dimensional spaces. We evaluate our approach with simulated users and a user study, demonstrating improved convergence, high success rates, and consistent gains over BO and line-search baselines, and further show the flexibility of the framework through several extensions.