# MOSAIC: Multi-Subject Personalized Generation via Correspondence-Aware Alignment and Disentanglement

arXiv: https://arxiv.org/abs/2509.01977 · HTML: https://mlsanity.com/paper/2509.01977/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 34 references that are in the map.
- First posted on arXiv 2 Sep 2025.

## Builds on (most cited first, 27 of 27)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [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)
- [FLUX.1 Kontext: Flow Matching for In-Context Image Generation and Editing in Latent Space](https://mlsanity.com/paper/2506.15742/) (2025)
- [ELLA: Equip Diffusion Models with LLM for Enhanced Semantic Alignment](https://mlsanity.com/paper/2403.05135/) (2024)
- [Emergent Correspondence from Image Diffusion](https://mlsanity.com/paper/2306.03881/) (2023)
- [BLIP-Diffusion: Pre-trained Subject Representation for Controllable Text-to-Image Generation and Editing](https://mlsanity.com/paper/2305.14720/) (2023)
- [OminiControl: Minimal and Universal Control for Diffusion Transformer](https://mlsanity.com/paper/2411.15098/) (2024)
- [A Tale of Two Features: Stable Diffusion Complements DINO for Zero-Shot Semantic Correspondence](https://mlsanity.com/paper/2305.15347/) (2023)
- [OmniGen2: Towards Instruction-Aligned Multimodal Generation](https://mlsanity.com/paper/2506.18871/) (2025)
- [Less-to-More Generalization: Unlocking More Controllability by In-Context Generation](https://mlsanity.com/paper/2504.02160/) (2025)
- [MS-Diffusion: Multi-subject Zero-shot Image Personalization with Layout Guidance](https://mlsanity.com/paper/2406.07209/) (2024)
- [SSR-Encoder: Encoding Selective Subject Representation for Subject-Driven Generation](https://mlsanity.com/paper/2312.16272/) (2023)
- [AP-10K: A Benchmark for Animal Pose Estimation in the Wild](https://mlsanity.com/paper/2108.12617/) (2021)
- [Telling Left from Right: Identifying Geometry-Aware Semantic Correspondence](https://mlsanity.com/paper/2311.17034/) (2023)
- [DreamO: A Unified Framework for Image Customization](https://mlsanity.com/paper/2504.16915/) (2025)
- [SFNet: Learning Object-aware Semantic Correspondence](https://mlsanity.com/paper/1904.01810/) (2019)
- [Proposal Flow](https://mlsanity.com/paper/1511.05065/) (2015)
- [Hyperpixel Flow: Semantic Correspondence with Multi-layer Neural Features](https://mlsanity.com/paper/1908.06537/) (2019)
- [XVerse: Consistent Multi-Subject Control of Identity and Semantic Attributes via DiT Modulation](https://mlsanity.com/paper/2506.21416/) (2025)
- [Reference-Based Sketch Image Colorization using Augmented-Self Reference and Dense Semantic Correspondence](https://mlsanity.com/paper/2005.05207/) (2020)
- [RegionDrag: Fast Region-Based Image Editing with Diffusion Models](https://mlsanity.com/paper/2407.18247/) (2024)

## Built on by (most cited first, 14 of 14)

- [PSR: Scaling Multi-Subject Personalized Image Generation with Pairwise Subject-Consistency Rewards](https://mlsanity.com/paper/2512.01236/) (2025)
- [OmniRefiner: Reinforcement-Guided Local Diffusion Refinement](https://mlsanity.com/paper/2511.19990/) (2025)
- [The Consistency Critic: Correcting Inconsistencies in Generated Images via Reference-Guided Attentive Alignment](https://mlsanity.com/paper/2511.20614/) (2025)
- [MultiBind: A Benchmark for Attribute Misbinding in Multi-Subject Generation](https://mlsanity.com/paper/2603.21937/) (2026)
- [CollectionLoRA: Collecting 50 Effects in 1 LoRA via Multi-Teacher On-Policy Distillation](https://mlsanity.com/paper/2605.25378/) (2026)
- [Scaling Multi-Reference Image Generation with Dynamic Reward Optimization](https://mlsanity.com/paper/2606.26947/) (2026)
- [When Identities Collapse: A Stress-Test Benchmark for Multi-Subject Personalization](https://mlsanity.com/paper/2603.26078/) (2026)
- [UniCustom: Unified Visual Conditioning for Multi-Reference Image Generation](https://mlsanity.com/paper/2605.12088/) (2026)
- [Hierarchical Concept-to-Appearance Guidance for Multi-Subject Image Generation](https://mlsanity.com/paper/2602.03448/) (2026)
- [Vanast: Virtual Try-On with Human Image Animation via Synthetic Triplet Supervision](https://mlsanity.com/paper/2604.04934/) (2026)
- [DreamShot: Personalized Storyboard Synthesis with Video Diffusion Prior](https://mlsanity.com/paper/2604.17195/) (2026)
- [CogCanvas: A Benchmark for Evaluating Multi-Subject Reference-Based Image Generation](https://mlsanity.com/paper/2606.15867/) (2026)
- [Training-Free Image Editing with Visual Context Integration and Concept Alignment](https://mlsanity.com/paper/2604.04487/) (2026)
- [Squeezing Capacity from Multimodal Large Language Models for Subject-driven Generation](https://mlsanity.com/paper/2605.26111/) (2026)

## Abstract (arXiv)

Multi-subject personalized generation presents unique challenges in maintaining identity fidelity and semantic coherence when synthesizing images conditioned on multiple reference subjects. Existing methods often suffer from identity blending and attribute leakage due to inadequate modeling of how different subjects should interact within shared representation spaces. We present MOSAIC, a representation-centric framework that rethinks multi-subject generation through explicit semantic correspondence and orthogonal feature disentanglement. Our key insight is that multi-subject generation requires precise semantic alignment at the representation level - knowing exactly which regions in the generated image should attend to which parts of each reference. To enable this, we introduce SemAlign-MS, a meticulously annotated dataset providing fine-grained semantic correspondences between multiple reference subjects and target images, previously unavailable in this domain. Building on this foundation, we propose the semantic correspondence attention loss to enforce precise point-to-point semantic alignment, ensuring high consistency from each reference to its designated regions. Furthermore, we develop the multi-reference disentanglement loss to push different subjects into orthogonal attention subspaces, preventing feature interference while preserving individual identity characteristics. Extensive experiments demonstrate that MOSAIC achieves state-of-the-art performance on multiple benchmarks. Notably, while existing methods typically degrade beyond 3 subjects, MOSAIC maintains high fidelity with 4+ reference subjects, opening new possibilities for complex multi-subject synthesis applications.