UniRef-Image-Edit: Towards Scalable and Consistent Multi-Reference Image Editing
Hongyang Wei, Bin Wen, Yancheng Long, Yankai Yang, Yuhang Hu, Tianke Zhang, Wei Chen, Haonan Fan, Kaiyu Jiang, Jiankang Chen, Changyi Liu, Kaiyu Tang, Haojie Ding, Xiao Yang, Jia Sun, Huaiqing Wang, Zhenyu Yang, Xinyu Wei, Xianglong He, Yangguang Li, Fan Yang, Tingting Gao, Lei Zhang, Guorui Zhou, Han Li
TL;DR
UniRef-Image-Edit is presented, a high-performance multi-modal generation system that unifies single-image editing and multi-image composition within a single framework and introduces Sequence-Extended Latent Fusion (SELF), a unified input representation that dynamically serializes multiple reference images into a coherent latent sequence.
Abstract
We present UniRef-Image-Edit, a high-performance multi-modal generation system that unifies single-image editing and multi-image composition within a single framework. Existing diffusion-based editing methods often struggle to maintain consistency across multiple conditions due to limited interaction between reference inputs. To address this, we introduce Sequence-Extended Latent Fusion (SELF), a unified input representation that dynamically serializes multiple reference images into a coherent latent sequence. During a dedicated training stage, all reference images are jointly constrained to fit within a fixed-length sequence under a global pixel-budget constraint. Building upon SELF, we propose a two-stage training framework comprising supervised fine-tuning (SFT) and reinforcement learning (RL). In the SFT stage, we jointly train on single-image editing and multi-image composition tasks to establish a robust generative prior. We adopt a progressive sequence length training strategy, in which all input images are initially resized to a total pixel budget of $1024^2$, and are then gradually increased to $1536^2$ and $2048^2$ to improve visual fidelity and cross-reference consistency. This gradual relaxation of compression enables the model to incrementally capture finer visual details while maintaining stable alignment across references. For the RL stage, we introduce Multi-Source GRPO (MSGRPO), to our knowledge the first reinforcement learning framework tailored for multi-reference image generation. MSGRPO optimizes the model to reconcile conflicting visual constraints, significantly enhancing compositional consistency. We will open-source the code, models, training data, and reward data for community research purposes.