# Contrastive Test-Time Composition of Multiple LoRA Models for Image Generation

arXiv: https://arxiv.org/abs/2403.19776 · HTML: https://mlsanity.com/paper/2403.19776/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Builds on 49 of its 59 references that are in the map.
- First posted on arXiv 28 Mar 2024.

## Builds on (most cited first, 30 of 49)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding](https://mlsanity.com/paper/2205.11487/) (2022)
- [A Style-Based Generator Architecture for Generative Adversarial Networks](https://mlsanity.com/paper/1812.04948/) (2018)
- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [Analyzing and Improving the Image Quality of StyleGAN](https://mlsanity.com/paper/1912.04958/) (2019)
- [DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation](https://mlsanity.com/paper/2208.12242/) (2022)
- [Perceptual Losses for Real-Time Style Transfer and Super-Resolution](https://mlsanity.com/paper/1603.08155/) (2016)
- [Prompt-to-Prompt Image Editing with Cross Attention Control](https://mlsanity.com/paper/2208.01626/) (2022)
- [An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversion](https://mlsanity.com/paper/2208.01618/) (2022)
- [Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization](https://mlsanity.com/paper/1703.06868/) (2017)
- [AnimateDiff: Animate Your Personalized Text-to-Image Diffusion Models without Specific Tuning](https://mlsanity.com/paper/2307.04725/) (2023)
- [RePaint: Inpainting using Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2201.09865/) (2022)
- [Multi-Concept Customization of Text-to-Image Diffusion](https://mlsanity.com/paper/2212.04488/) (2022)
- [Blended Diffusion for Text-driven Editing of Natural Images](https://mlsanity.com/paper/2111.14818/) (2021)
- [Attend-and-Excite: Attention-Based Semantic Guidance for Text-to-Image Diffusion Models](https://mlsanity.com/paper/2301.13826/) (2023)
- [DiffEdit: Diffusion-based semantic image editing with mask guidance](https://mlsanity.com/paper/2210.11427/) (2022)
- [Segment Everything Everywhere All at Once](https://mlsanity.com/paper/2304.06718/) (2023)
- [MultiDiffusion: Fusing Diffusion Paths for Controlled Image Generation](https://mlsanity.com/paper/2302.08113/) (2023)
- [Blended Latent Diffusion](https://mlsanity.com/paper/2206.02779/) (2022)
- [LoraHub: Efficient Cross-Task Generalization via Dynamic LoRA Composition](https://mlsanity.com/paper/2307.13269/) (2023)
- [DiffusionDet: Diffusion Model for Object Detection](https://mlsanity.com/paper/2211.09788/) (2022)
- [Composer: Creative and Controllable Image Synthesis with Composable Conditions](https://mlsanity.com/paper/2302.09778/) (2023)

## Built on by (most cited first, 10 of 10)

- [MotionFlow: Attention-Driven Motion Transfer in Video Diffusion Models](https://mlsanity.com/paper/2412.05275/) (2024)
- [LoRAShop: Training-Free Multi-Concept Image Generation and Editing with Rectified Flow Transformers](https://mlsanity.com/paper/2505.23758/) (2025)
- [FreeLoRA: Enabling Training-Free LoRA Fusion for Autoregressive Multi-Subject Personalization](https://mlsanity.com/paper/2507.01792/) (2025)
- [SplitFlux: Learning to Decouple Content and Style from a Single Image](https://mlsanity.com/paper/2511.15258/) (2025)
- [Compression as Adaptation: Implicit Visual Representation with Diffusion Foundation Models](https://mlsanity.com/paper/2603.07615/) (2026)
- [Not All Layers Are Created Equal: Adaptive LoRA Ranks for Personalized Image Generation](https://mlsanity.com/paper/2603.21884/) (2026)
- [Training-Free Multi-Concept LoRA Composition with Prompt-Aware Weighting](https://mlsanity.com/paper/2606.03792/) (2026)
- [CopyCat: Improving Fine-Grained Subject Consistency in Subject-to-Image Models within Seconds](https://mlsanity.com/paper/2608.00674/) (2026)
- [InstructionCrafter: Generating Consistent and High-Fidelity Visual Instructions](https://mlsanity.com/paper/2608.08460/) (2026)
- [LoRango: It Takes Two LoRAs to Unlock Hidden Behaviors in Diffusion Models](https://mlsanity.com/paper/2609.25884/) (2026)

## Abstract (arXiv)

Low-Rank Adaptation (LoRA) has emerged as a powerful and popular technique for personalization, enabling efficient adaptation of pre-trained image generation models for specific tasks without comprehensive retraining. While employing individual pre-trained LoRA models excels at representing single concepts, such as those representing a specific dog or a cat, utilizing multiple LoRA models to capture a variety of concepts in a single image still poses a significant challenge. Existing methods often fall short, primarily because the attention mechanisms within different LoRA models overlap, leading to scenarios where one concept may be completely ignored (e.g., omitting the dog) or where concepts are incorrectly combined (e.g., producing an image of two cats instead of one cat and one dog). We introduce CLoRA, a training-free approach that addresses these limitations by updating the attention maps of multiple LoRA models at test-time, and leveraging the attention maps to create semantic masks for fusing latent representations. This enables the generation of composite images that accurately reflect the characteristics of each LoRA. Our comprehensive qualitative and quantitative evaluations demonstrate that CLoRA significantly outperforms existing methods in multi-concept image generation using LoRAs.