Related papers: ChimeraLoRA: Multi-Head LoRA-Guided Synthetic Datasets

ChimeraLoRA: Multi-Head LoRA-Guided Synthetic Datasets

URL: http://arxiv.org/abs/2602.19708v1
Date: Mon, 23 Feb 2026 10:59:41 GMT
Title: ChimeraLoRA: Multi-Head LoRA-Guided Synthetic Datasets
Authors: Hoyoung Kim, Minwoo Jang, Jabin Koo, Sangdoo Yun, Jungseul Ok,
Abstract summary: We propose a class-wise LoRA trained on few-shot real sets to synthesize additional images.<n>Across diverse datasets, our synthesized images are both diverse and detail-rich while closely aligning with the few-shot real distribution.
Score: 48.20200154292035
License: http://creativecommons.org/licenses/by/4.0/
Abstract: Beyond general recognition tasks, specialized domains including privacy-constrained medical applications and fine-grained settings often encounter data scarcity, especially for tail classes. To obtain less biased and more reliable models under such scarcity, practitioners leverage diffusion models to supplement underrepresented regions of real data. Specifically, recent studies fine-tune pretrained diffusion models with LoRA on few-shot real sets to synthesize additional images. While an image-wise LoRA trained on a single image captures fine-grained details yet offers limited diversity, a class-wise LoRA trained over all shots produces diverse images as it encodes class priors yet tends to overlook fine details. To combine both benefits, we separate the adapter into a class-shared LoRA~$A$ for class priors and per-image LoRAs~$\mathcal{B}$ for image-specific characteristics. To expose coherent class semantics in the shared LoRA~$A$, we propose a semantic boosting by preserving class bounding boxes during training. For generation, we compose $A$ with a mixture of $\mathcal{B}$ using coefficients drawn from a Dirichlet distribution. Across diverse datasets, our synthesized images are both diverse and detail-rich while closely aligning with the few-shot real distribution, yielding robust gains in downstream classification accuracy.

Related papers

FreeFuse: Multi-Subject LoRA Fusion via Auto Masking at Test Time [22.751943468067633]
FreeFuse is a training-free approach for multi-subject text-to-image generation through automatic fusion of multiple subject LoRAs.<n>It demonstrates superior practicality and efficiency as it requires no additional training, no modification to LoRAs, no auxiliary models, and no user-defined prompt templates or region specifications.
arXiv Detail & Related papers (2025-10-27T16:54:08Z)
LoFT: LoRA-fused Training Dataset Generation with Few-shot Guidance [96.6544564242316]
We introduce a novel dataset generation framework named LoFT, LoRA-Fused Training-data Generation with Few-shot Guidance.<n>Our method fine-tunes LoRA weights on individual real images and fuses them at inference time, producing synthetic images that combine the features of real images for improved diversity and fidelity of generated data.<n>Our experiments show that training on LoFT-generated data consistently outperforms other synthetic dataset methods, significantly increasing accuracy as the dataset size increases.
arXiv Detail & Related papers (2025-05-16T21:17:55Z)
Cached Multi-Lora Composition for Multi-Concept Image Generation [10.433033595844442]
Low-Rank Adaptation (LoRA) has emerged as a widely adopted technique in text-to-image models.<n>Current approaches face significant challenges when composing these LoRAs for multi-concept image generation.<n>We introduce a novel, training-free framework, Cached Multi-LoRA (CMLoRA), designed to efficiently integrate multiple LoRAs.
arXiv Detail & Related papers (2025-02-07T13:41:51Z)
A LoRA is Worth a Thousand Pictures [28.928964530616593]
Low Rank Adaptation (LoRA) can replicate an artist's style or subject using minimal data and computation.<n>We show that LoRA weights alone can serve as an effective descriptor of style, without the need for additional image generation or knowledge of the original training set.<n>We conclude with a discussion on potential future applications, such as zero-shot LoRA fine-tuning and model attribution.
arXiv Detail & Related papers (2024-12-16T18:18:17Z)
LoRA of Change: Learning to Generate LoRA for the Editing Instruction from A Single Before-After Image Pair [116.48684498656871]
We propose the LoRA of Change (LoC) framework for image editing with visual instructions, i.e., before-after image pairs.<n>We learn an instruction-specific LoRA to encode the "change" in a before-after image pair, enhancing the interpretability and reusability of our model.<n>Our model produces high-quality images that align with user intent and support a broad spectrum of real-world visual instructions.
arXiv Detail & Related papers (2024-11-28T13:55:06Z)
Contrastive Test-Time Composition of Multiple LoRA Models for Image Generation [44.037664077117945]
Low-Rank Adaptation (LoRA) has emerged as a powerful and popular technique for personalization.<n>Existing methods often fall short because the attention mechanisms within different LoRA models overlap.<n>We introduce CLoRA, a training-free approach that addresses these limitations by updating the attention maps of multiple LoRA models at test-time.
arXiv Detail & Related papers (2024-03-28T18:58:43Z)
Multi-LoRA Composition for Image Generation [107.83002438126832]
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.
arXiv Detail & Related papers (2024-02-26T18:59:18Z)
Continual Diffusion: Continual Customization of Text-to-Image Diffusion with C-LoRA [64.10981296843609]
We show that recent state-of-the-art customization of text-to-image models suffer from catastrophic forgetting when new concepts arrive sequentially. We propose a new method, C-LoRA, composed of a continually self-regularized low-rank adaptation in cross attention layers of the popular Stable Diffusion model. We show that C-LoRA not only outperforms several baselines for our proposed setting of text-to-image continual customization, but that we achieve a new state-of-the-art in the well-established rehearsal-free continual learning setting for image classification.
arXiv Detail & Related papers (2023-04-12T17:59:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.