# USO: Unified Style and Subject-Driven Generation via Disentangled and Reward Learning

arXiv: https://arxiv.org/abs/2508.18966 · HTML: https://mlsanity.com/paper/2508.18966/

## Key facts

- Cited by 27 papers in the map (data through 14 Oct 2026).
- Builds on 40 of its 41 references that are in the map.
- First posted on arXiv 26 Aug 2025.

## Builds on (most cited first, 30 of 40)

- [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)
- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [Scaling Rectified Flow Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.03206/) (2024)
- [DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation](https://mlsanity.com/paper/2208.12242/) (2022)
- [Sigmoid Loss for Language Image Pre-Training](https://mlsanity.com/paper/2303.15343/) (2023)
- [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)
- [IP-Adapter: Text Compatible Image Prompt Adapter for Text-to-Image Diffusion Models](https://mlsanity.com/paper/2308.06721/) (2023)
- [ImageReward: Learning and Evaluating Human Preferences for Text-to-Image Generation](https://mlsanity.com/paper/2304.05977/) (2023)
- [Qwen-Image Technical Report](https://mlsanity.com/paper/2508.02324/) (2025)
- [FLUX.1 Kontext: Flow Matching for In-Context Image Generation and Editing in Latent Space](https://mlsanity.com/paper/2506.15742/) (2025)
- [Emerging Properties in Unified Multimodal Pretraining](https://mlsanity.com/paper/2505.14683/) (2025)
- [BLIP-Diffusion: Pre-trained Subject Representation for Controllable Text-to-Image Generation and Editing](https://mlsanity.com/paper/2305.14720/) (2023)
- [ELITE: Encoding Visual Concepts into Textual Embeddings for Customized Text-to-Image Generation](https://mlsanity.com/paper/2302.13848/) (2023)
- [OmniGen: Unified Image Generation](https://mlsanity.com/paper/2409.11340/) (2024)
- [OminiControl: Minimal and Universal Control for Diffusion Transformer](https://mlsanity.com/paper/2411.15098/) (2024)
- [OmniGen2: Towards Instruction-Aligned Multimodal Generation](https://mlsanity.com/paper/2506.18871/) (2025)
- [Re-Imagen: Retrieval-Augmented Text-to-Image Generator](https://mlsanity.com/paper/2209.14491/) (2022)
- [Style Injection in Diffusion: A Training-free Approach for Adapting Large-scale Diffusion Models for Style Transfer](https://mlsanity.com/paper/2312.09008/) (2023)
- [InstantStyle: Free Lunch towards Style-Preserving in Text-to-Image Generation](https://mlsanity.com/paper/2404.02733/) (2024)
- [In-Context LoRA for Diffusion Transformers](https://mlsanity.com/paper/2410.23775/) (2024)
- [Less-to-More Generalization: Unlocking More Controllability by In-Context Generation](https://mlsanity.com/paper/2504.02160/) (2025)
- [SSR-Encoder: Encoding Selective Subject Representation for Subject-Driven Generation](https://mlsanity.com/paper/2312.16272/) (2023)
- [Implicit Style-Content Separation using B-LoRA](https://mlsanity.com/paper/2403.14572/) (2024)
- [Measuring Style Similarity in Diffusion Models](https://mlsanity.com/paper/2404.01292/) (2024)
- [DreamO: A Unified Framework for Image Customization](https://mlsanity.com/paper/2504.16915/) (2025)
- [CSGO: Content-Style Composition in Text-to-Image Generation](https://mlsanity.com/paper/2408.16766/) (2024)
- [DEADiff: An Efficient Stylization Diffusion Model with Disentangled Representations](https://mlsanity.com/paper/2403.06951/) (2024)
- [Visual Style Prompting with Swapping Self-Attention](https://mlsanity.com/paper/2402.12974/) (2024)

## Built on by (most cited first, 27 of 27)

- [WithAnyone: Towards Controllable and ID Consistent Image Generation](https://mlsanity.com/paper/2510.14975/) (2025)
- [Lance: Unified Multimodal Modeling by Multi-Task Synergy](https://mlsanity.com/paper/2605.18678/) (2026)
- [Scone: Bridging Composition and Distinction in Subject-Driven Image Generation via Unified Understanding-Generation Modeling](https://mlsanity.com/paper/2512.12675/) (2025)
- [EditIDv2: Editable ID Customization with Data-Lubricated ID Feature Integration for Text-to-Image Generation](https://mlsanity.com/paper/2509.05659/) (2025)
- [iMontage: Unified, Versatile, Highly Dynamic Many-to-many Image Generation](https://mlsanity.com/paper/2511.20635/) (2025)
- [InnoAds-Composer: Efficient Condition Composition for E-Commerce Poster Generation](https://mlsanity.com/paper/2603.05898/) (2026)
- [3SGen: Unified Subject, Style, and Structure-Driven Image Generation with Adaptive Task-specific Memory](https://mlsanity.com/paper/2512.19271/) (2025)
- [Loom: Diffusion-Transformer for Interleaved Generation](https://mlsanity.com/paper/2512.18254/) (2025)
- [Garments2Look: A Multi-Reference Dataset for High-Fidelity Outfit-Level Virtual Try-On with Clothing and Accessories](https://mlsanity.com/paper/2603.14153/) (2026)
- [EmoStyle: Emotion-Driven Image Stylization](https://mlsanity.com/paper/2512.05478/) (2025)
- [UniRef-Image-Edit: Towards Scalable and Consistent Multi-Reference Image Editing](https://mlsanity.com/paper/2602.14186/) (2026)
- [Scaling Multi-Reference Image Generation with Dynamic Reward Optimization](https://mlsanity.com/paper/2606.26947/) (2026)
- [UniCustom: Unified Visual Conditioning for Multi-Reference Image Generation](https://mlsanity.com/paper/2605.12088/) (2026)
- [OmniStyle2: Learning to Stylize by Learning to Destylize](https://mlsanity.com/paper/2509.05970/) (2025)
- [A Style is Worth One Code: Unlocking Code-to-Style Image Generation with Discrete Style Space](https://mlsanity.com/paper/2511.10555/) (2025)
- [Towards Generalized Multi-Image Editing for Unified Multimodal Models](https://mlsanity.com/paper/2601.05572/) (2026)
- [Mixture of Style Experts for Diverse Image Stylization](https://mlsanity.com/paper/2603.16649/) (2026)
- [FreeStyle: Free Control of Style-Content Dual-Reference Generation from Community LoRA Mining](https://mlsanity.com/paper/2606.20506/) (2026)
- [Unleashing Multimodal Large Language Models for Training-free HOI Detection in the Wild](https://mlsanity.com/paper/2607.13881/) (2026)
- [DreamStyle3D: Efficient 3D Stylized Asset Generation via Dual-Attention Disentanglement](https://mlsanity.com/paper/2607.24721/) (2026)
- [AnyPhoto: Multi-Person Identity Preserving Image Generation with ID Adaptive Modulation on Location Canvas](https://mlsanity.com/paper/2603.14770/) (2026)
- [UniCSG: Unified High-Fidelity Content-Constrained Style-Driven Generation via Staged Semantic and Frequency Disentanglement](https://mlsanity.com/paper/2604.17850/) (2026)
- [Fashion130K: An E-commerce Fashion Dataset for Outfit Generation with Unified Multi-modal Condition](https://mlsanity.com/paper/2605.10127/) (2026)
- [Squeezing Capacity from Multimodal Large Language Models for Subject-driven Generation](https://mlsanity.com/paper/2605.26111/) (2026)
- [CopyCat: Improving Fine-Grained Subject Consistency in Subject-to-Image Models within Seconds](https://mlsanity.com/paper/2608.00674/) (2026)
- [CRAFT: Constrained Reward via Attention Fine-Tuning for Subject Personalization without Composed Targets](https://mlsanity.com/paper/2608.14403/) (2026)
- [WithEveryone: Unified Planning and Identity Grounding for Group Image Generation](https://mlsanity.com/paper/2608.20336/) (2026)

## Abstract (arXiv)

Existing literature typically treats style-driven and subject-driven generation as two disjoint tasks: the former prioritizes stylistic similarity, whereas the latter insists on subject consistency, resulting in an apparent antagonism. We argue that both objectives can be unified under a single framework because they ultimately concern the disentanglement and re-composition of content and style, a long-standing theme in style-driven research. To this end, we present USO, a Unified Style-Subject Optimized customization model. First, we construct a large-scale triplet dataset consisting of content images, style images, and their corresponding stylized content images. Second, we introduce a disentangled learning scheme that simultaneously aligns style features and disentangles content from style through two complementary objectives, style-alignment training and content-style disentanglement training. Third, we incorporate a style reward-learning paradigm denoted as SRL to further enhance the model's performance. Finally, we release USO-Bench, the first benchmark that jointly evaluates style similarity and subject fidelity across multiple metrics. Extensive experiments demonstrate that USO achieves state-of-the-art performance among open-source models along both dimensions of subject consistency and style similarity. Code and model: https://github.com/bytedance/USO