arXiv · 2611.22969
Rescaling OLS-Merged Adapters To Improve Personalized Image Generation
Abstract
Merging multiple adapters into a single image generation model reduces deployment costs while preserving diverse task capabilities or personalization modes. State-of-the-art methods such as RegMean, Mix-of-Show, and Subspace Routing fit the outputs of the merged model to those of the individual adapters using ordinary least squares (OLS). We show that this reconstruction objective systematically shrinks task-specific signals whenever the activation distributions of different tasks overlap, producing a merged adapter that is too weak for downstream use. We derive a decomposition that quantifies this shrinkage and propose rescaling the merged update to restore signal strength. On multi-subject personalization with up to 30 merged adapters, rescaling substantially improves subject fidelity over unscaled baselines. Our approach also improves results on style transfer, concept erasure, image classification, and LLM-based text classification.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Leander Girrbach, Niklas Schweiger, Zeynep Akata. 2026-11-17. Rescaling OLS-Merged Adapters To Improve Personalized Image Generation. https://searcharxiv.com/abs/2611.22969
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Searcharxiv by Indiciae: BibTeX · RIS.