# Scaling Multi-Reference Image Generation with Dynamic Reward Optimization

arXiv: https://arxiv.org/abs/2606.26947 · HTML: https://mlsanity.com/paper/2606.26947/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 35 of its 64 references that are in the map.
- First posted on arXiv 25 Jun 2026.

## Builds on (most cited first, 30 of 35)

- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Auto-Encoding Variational Bayes](https://mlsanity.com/paper/1312.6114/) (2013)
- [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://mlsanity.com/paper/2303.05499/) (2023)
- [SAM 2: Segment Anything in Images and Videos](https://mlsanity.com/paper/2408.00714/) (2024)
- [Decoupled Weight Decay Regularization](https://mlsanity.com/paper/1711.05101/) (2017)
- [IP-Adapter: Text Compatible Image Prompt Adapter for Text-to-Image Diffusion Models](https://mlsanity.com/paper/2308.06721/) (2023)
- [SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features](https://mlsanity.com/paper/2502.14786/) (2025)
- [Qwen-Image Technical Report](https://mlsanity.com/paper/2508.02324/) (2025)
- [Emerging Properties in Unified Multimodal Pretraining](https://mlsanity.com/paper/2505.14683/) (2025)
- [Gemini: A Family of Highly Capable Multimodal Models](https://mlsanity.com/paper/2312.11805/) (2023)
- [DeepSeek-V3 Technical Report](https://mlsanity.com/paper/2412.19437/) (2024)
- [Flow-GRPO: Training Flow Matching Models via Online RL](https://mlsanity.com/paper/2505.05470/) (2025)
- [Step1X-Edit: A Practical Framework for General Image Editing](https://mlsanity.com/paper/2504.17761/) (2025)
- [OmniGen2: Towards Instruction-Aligned Multimodal Generation](https://mlsanity.com/paper/2506.18871/) (2025)
- [Seedream 4.0: Toward Next-generation Multimodal Image Generation](https://mlsanity.com/paper/2509.20427/) (2025)
- [ImgEdit: A Unified Image Editing Dataset and Benchmark](https://mlsanity.com/paper/2505.20275/) (2025)
- [In-Context Edit: Enabling Instructional Image Editing with In-Context Generation in Large Scale Diffusion Transformer](https://mlsanity.com/paper/2504.20690/) (2025)
- [GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization](https://mlsanity.com/paper/2601.05242/) (2026)
- [Measuring Style Similarity in Diffusion Models](https://mlsanity.com/paper/2404.01292/) (2024)
- [XVerse: Consistent Multi-Subject Control of Identity and Semantic Attributes via DiT Modulation](https://mlsanity.com/paper/2506.21416/) (2025)
- [Skywork UniPic 2.0: Building Kontext Model with Online RL for Unified Multimodal Model](https://mlsanity.com/paper/2509.04548/) (2025)
- [DreamOmni2: Multimodal Instruction-based Editing and Generation](https://mlsanity.com/paper/2510.06679/) (2025)
- [USO: Unified Style and Subject-Driven Generation via Disentangled and Reward Learning](https://mlsanity.com/paper/2508.18966/) (2025)
- [UMO: Scaling Multi-Identity Consistency for Image Customization via Matching Reward](https://mlsanity.com/paper/2509.06818/) (2025)
- [ContextGen: Contextual Layout Anchoring for Identity-Consistent Multi-Instance Generation](https://mlsanity.com/paper/2510.11000/) (2025)
- [MOSAIC: Multi-Subject Personalized Generation via Correspondence-Aware Alignment and Disentanglement](https://mlsanity.com/paper/2509.01977/) (2025)
- [Flow-Factory: A Unified Framework for Reinforcement Learning in Flow-Matching Models](https://mlsanity.com/paper/2602.12529/) (2026)
- [MultiBanana: A Challenging Benchmark for Multi-Reference Text-to-Image Generation](https://mlsanity.com/paper/2511.22989/) (2025)

## Built on by (most cited first, 3 of 3)

- [SenseNova-U1.5: Towards Native Unified Visual Intelligence](https://mlsanity.com/paper/2609.11929/) (2026)
- [AutoRef: Harness Optimization for Agentic Multi-Reference Image Generation](https://mlsanity.com/paper/2609.35530/) (2026)
- [VIF-Bench: Evaluating Visual Instruction Following in Multi-Reference Image Generation](https://mlsanity.com/paper/2609.37709/) (2026)

## Abstract (arXiv)

While personalized image generation has achieved remarkable progress, multi-reference image generation (MRIG) remains a challenging task. Most existing benchmarks fail to adequately evaluate complex MRIG scenarios, hindering further progress in this area. To better assess model performance on complex MRIG tasks, we introduce OmniRef-Bench, a benchmark that covers complex combinations of reference image types and a large number of reference images. Evaluations on OmniRef-Bench show that mainstream open-source models struggle in complex MRIG scenarios, and their performance deteriorates significantly as the number of mixed-type reference images increases. To address this issue, we propose DyRef, a two-stage training framework. In the first stage, supervised fine-tuning equips the model with the basic capability to handle complex MRIG tasks. In the second stage, we introduce Difficulty-aware Advantage Reweighting (DAR) and Discriminative Reward Scaling (DRS). DAR dynamically adjusts the optimization objective to improve performance when handling a large number of mixed-type reference images. DRS enlarges intra-group reward differences for more effective policy optimization. Experiments demonstrate that DyRef significantly improves the performance of open-source models on OmniRef-Bench and single-image editing benchmarks, demonstrating the effectiveness and generalization capability of our approach.