Training-Free Multi-Concept Image Editing

TL;DR

This paper introduces a training-free, unified framework called Concept Distillation Sampling (CDS) for multi-concept image editing that preserves identity and details without reference samples, outperforming existing methods.

Contribution

We propose the first training-free, target-less multi-concept image editing framework that integrates a stable distillation backbone with dynamic weighting for seamless concept composition.

Findings

Achieves state-of-the-art results on InstructPix2Pix and ComposLoRA benchmarks.

Preserves identity and intricate details without reference images.

Outperforms existing training-free and multi-LoRA methods.

Abstract

Editing images with diffusion models under strict training-free constraints remains a significant challenge. While recent optimisation-based methods achieve strong zero-shot edits from text, they struggle to preserve identity and capture intricate details, such as facial structure, material texture, or object-specific geometry, that exist below the level of linguistic abstraction. To address this fundamental gap, we propose Concept Distillation Sampling (CDS). To the best of our knowledge, we are the first to introduce a unified, training-free framework for target-less, multi-concept image editing. CDS overcomes the linguistic bottleneck of previous methods by integrating a highly stable distillation backbone (featuring ordered timesteps, regularisation, and negative-prompt guidance), with a dynamic weighting mechanism. This approach enables the seamless composition and control of multiple visual concepts directly within the diffusion process, utilising spatially-aware priors from pretrained LoRA adapters without spatial interference. Our method preserves instance fidelity without requiring reference samples of the desired edit. Extensive quantitative and qualitative evaluations demonstrate consistent state-of-the-art performance over existing training-free editing and multi-LoRA composition methods on the InstructPix2Pix and ComposLoRA benchmarks. Code will be made publicly available.