mlsanityData to 14 Oct 2026

Multi-LoRA Composition for Image Generation

arXiv 2402.16843 ↗First posted 26 Feb 2024cs.CV cs.AI cs.CLcited by 80

Not on the map. Why?

Cited by 25 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Low-Rank Adaptation (LoRA) is extensively utilized in text-to-image models for the accurate rendition of specific elements like distinct characters or unique styles in generated images. Nonetheless, existing methods face challenges in effectively composing multiple LoRAs, especially as the number of LoRAs to be integrated grows, thus hindering the creation of complex imagery. In this paper, we study multi-LoRA composition through a decoding-centric perspective. We present two training-free methods: LoRA Switch, which alternates between different LoRAs at each denoising step, and LoRA Composite, which simultaneously incorporates all LoRAs to guide more cohesive image synthesis. To evaluate the proposed approaches, we establish ComposLoRA, a new comprehensive testbed as part of this research. It features a diverse range of LoRA categories with 480 composition sets. Utilizing an evaluation framework based on GPT-4V, our findings demonstrate a clear improvement in performance with our methods over the prevalent baseline, particularly evident when increasing the number of LoRAs in a composition. The code, benchmarks, LoRA weights, and all evaluation details are available on our project website: https://maszhongming.github.io/Multi-LoRA-Composition.

Key facts. Cited by 80 papers in the map (data through 14 Oct 2026). Builds on 34 of its 41 references that are in the map. First posted on arXiv 26 Feb 2024.

Builds on

34 of 41 references.

References · 34

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  3. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 9,122
  4. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,255
  5. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,058
  6. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,585
  7. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,355
  8. 2022Classifier-Free Diffusion Guidancecited by 5,847
  9. 2021GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Modelscited by 3,614
  10. 2022DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generationcited by 3,300
  11. 2021CLIPScore: A Reference-free Evaluation Metric for Image Captioningcited by 2,424
  12. 2022DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Stepscited by 1,979
22 more
  1. 2022DPM-Solver++: Fast Solver for Guided Sampling of Diffusion Probabilistic Modelscited by 938
  2. 2020Intrinsic Dimensionality Explains the Effectiveness of Language Model Fine-Tuningcited by 671
  3. 2022Compositional Visual Generation with Composable Diffusion Modelscited by 624
  4. 2022Make-A-Scene: Scene-Based Text-to-Image Generation with Human Priorscited by 487
  5. 2022Training-Free Structured Diffusion Guidance for Compositional Text-to-Image Synthesiscited by 388
  6. 2023LoraHub: Efficient Cross-Task Generalization via Dynamic LoRA Compositioncited by 364
  7. 2018Image Generation from Scene Graphscited by 324
  8. 2023Reduce, Reuse, Recycle: Compositional Generation with Energy-Based Diffusion Models and MCMCcited by 220
  9. 2023StyleDrop: Text-to-Image Generation in Any Stylecited by 199
  10. 2023ZipLoRA: Any Subject in Any Style by Effectively Merging LoRAscited by 194
  11. 2023Composing Parameter-Efficient Modules with Arithmetic Operationscited by 156
  12. 2023GPT-4V(ision) as a Generalist Evaluator for Vision-Language Taskscited by 119
  13. 2023Collaborative Diffusion for Multi-Modal Face Generation and Editingcited by 116
  14. 2023In-Context Learning Unlocked for Diffusion Modelscited by 104
  15. 2023ImagenHub: Standardizing the evaluation of conditional image generation modelscited by 77
  16. 2021Learning to Compose Visual Relationscited by 72
  17. 2021Controllable and Compositional Generation with Latent-Space Energy-Based Modelscited by 64
  18. 2022Modeling Image Composition for Complex Scene Generationcited by 43
  19. 2023Unsupervised Compositional Concepts Discovery with Text-to-Image Generative Modelscited by 36
  20. 2022Composing Ensembles of Pre-trained Models via Iterative Consensuscited by 28
  21. 2023Seeking Neural Nuggets: Knowledge Transfer in Large Language Models from a Parametric Perspectivecited by 17
  22. 2023DEsignBench: Exploring and Benchmarking DALL-E 3 for Imagining Visual Designcited by 17

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 26; 2025: 31; 2026: 23.

80 papers.

Sort:

By area:

  1. 2024ReFT: Representation Finetuning for Language Modelscited by 197
  2. 2024LoRA-Composer: Leveraging Low-Rank Adaptation for Multi-Concept Customization in Training-Free Diffusion Modelscited by 55
  3. 2024LoRA Soups: Merging LoRAs for Practical Skill Composition Taskscited by 52
  4. 2025K-LoRA: Unlocking Training-Free Fusion of Any Subject and Style LoRAscited by 48
  5. 2024HiCo: Hierarchical Controllable Diffusion Model for Layout-to-image Generationcited by 41
  6. 2024Low-Rank Adaptation for Foundation Models: A Comprehensive Reviewcited by 37
  7. 2024The Superposition of Diffusion Models Using the Ito Density Estimatorcited by 36
  8. 2024How to Continually Adapt Text-to-Image Diffusion Models for Flexible Customization?cited by 33
  9. 2024Getting it Right: Improving Spatial Consistency in Text-to-Image Modelscited by 30
  10. 2024MC2: Multi-concept Guidance for Customized Multi-concept Generationcited by 26
  11. 2024AlignGuard: Scalable Safety Alignment for Text-to-Image Generationcited by 20
  12. 2024FouRA: Fourier Low Rank Adaptationcited by 15
38 more
  1. 2024SELMA: Learning and Merging Skill-Specific Text-to-Image Experts with Auto-Generated Datacited by 14
  2. 2024Retrieval-Augmented Mixture of LoRA Experts for Uploadable Machine Learningcited by 14
  3. 2025Skill Expansion and Composition in Parameter Spacecited by 14
  4. 2024Realistic Evaluation of Model Merging for Compositional Generalizationcited by 12
  5. 2024IterIS: Iterative Inference-Solving Alignment for LoRA Mergingcited by 11
  6. 2024Contrastive Test-Time Composition of Multiple LoRA Models for Image Generationcited by 10
  7. 20243-in-1: 2D Rotary Adaptation for Efficient Finetuning, Efficient Batching and Composabilitycited by 10
  8. 2024S2FT: Efficient, Scalable and Generalizable LLM Fine-tuning by Structured Sparsitycited by 10
  9. 2025Towards Higher Effective Rank in Parameter-efficient Fine-tuning using Khatri--Rao Productcited by 10
  10. 2025EigenLoRAx: Recycling Adapters to Find Principal Subspaces for Resource-Efficient Adaptation and Inferencecited by 9
  11. 2025QR-LoRA: Efficient and Disentangled Fine-tuning via QR Decomposition for Customized Generationcited by 9
  12. 2024Enhancing Adversarial Robustness of Vision-Language Models through Low-Rank Adaptationcited by 8
  13. 2024DreamRunner: Fine-Grained Compositional Story-to-Video Generation with Retrieval-Augmented Motion Adaptationcited by 7
  14. 2025Diffusion Blend: Inference-Time Multi-Preference Alignment for Diffusion Modelscited by 7
  15. 2025LiON-LoRA: Rethinking LoRA Fusion to Unify Controllable Spatial and Temporal Generation for Video Diffusioncited by 7
  16. 2025Training-Free Multi-Style Fusion Through Reference-Based Adaptive Modulationcited by 7
  17. 2024Can video generation replace cinematographers? Research on the cinematic language of generated videocited by 6
  18. 2025Fine Tuning without Catastrophic Forgetting via Selective Low Rank Adaptationcited by 6
  19. 2025MultLFG: Training-free Multi-LoRA composition using Frequency-domain Guidancecited by 6
  20. 2025DyME: Dynamic Multi-Concept Erasure in Diffusion Models with Bi-Level Orthogonal LoRA Adaptationcited by 6
  21. 2025FreeLoRA: Enabling Training-Free LoRA Fusion for Autoregressive Multi-Subject Personalizationcited by 5
  22. 2025SplitFlux: Learning to Decouple Content and Style from a Single Imagecited by 5
  23. 2026ReStyle-TTS: Relative and Continuous Style Control for Zero-Shot Speech Synthesiscited by 5
  24. 2024Training-free Color-Style Disentanglement for Constrained Text-to-Image Synthesiscited by 4
  25. 2024Modification Takes Courage: Seamless Image Stitching via Reference-Driven Inpaintingcited by 4
  26. 2024Safety Alignment Backfires: Preventing the Re-emergence of Suppressed Concepts in Fine-tuned Text-to-Image Diffusion Modelscited by 4
  27. 2025PAVE: Patching and Adapting Video Large Language Modelscited by 4
  28. 2025Distribution-Conditional Generation: From Class Distribution to Creative Generationcited by 4
  29. 2026Slot-ID: Identity-Preserving Video Generation from Reference Videos via Slot-Based Temporal Identity Encodingcited by 4
  30. 2026DiffGraph: An Automated Agent-driven Model Merging Framework for In-the-Wild Text-to-Image Generationcited by 4
  31. 2026Label-Free Cross-Task LoRA Merging with Null-Space Compressioncited by 4
  32. 2024Multi LoRA Meets Vision: Merging multiple adapters to create a multi task modelcited by 3
  33. 2025FLoRA: Sample-Efficient Preference-based RL via Low-Rank Style Adaptation of Reward Functionscited by 3
  34. 2025LoRAtorio: An intrinsic approach to LoRA Skill Compositioncited by 3
  35. 2025Rethinking Inter-LoRA Orthogonality in Adapter Merging: Insights from Orthogonal Monte Carlo Dropoutcited by 3
  36. 2025LoRAverse: A Submodular Framework to Retrieve Diverse Adapters for Diffusion Modelscited by 3
  37. 2025MapReduce LoRA: Advancing the Pareto Front in Multi-Preference Optimization for Generative Modelscited by 3
  38. 2026Learning to Compose for Cross-domain Agentic Workflow Generationcited by 3

50 most cited of 80 All 80 (JSON)