The paper introduces ARRO, a reinforcement‑learning framework that enforces minimal‑change editing by auditing source images, instructions, and edited outputs for unimplemented and unintended changes. Using a vision‑language reward model and a group‑level rubric, ARRO achieves higher EditScore on multiple benchmarks and reduces off‑target pixel changes by 8.4% on FLUX.1 Kontext‑dev. The approach eliminates the need for per‑instruction human annotations and demonstrates strong performance across several evaluation sets.
RefineEdit is a training‑free prompt‑to‑prompt image editing framework that uses a Generative Refinement Network to edit images by refining binary image codes. It couples edit localization with content generation, selecting editable positions based on signed probability differences between an editing branch and a source branch, and stabilizes edits with adaptive spatial freezing and finite bit locking. The method requires no additional training, external masks, or attention control, and outperforms other methods on PIE‑Bench in background‑preservation metrics and CLIP scores.
By Yulong Chen, Ziqian Zhang, Haoyu Zhang, Ao He, Senmao Li, Kai Wang
arXiv:2605.08703v2 Announce Type: replace
Abstract: Aligning generative models with human preferences remains a central challenge. While RLHF is widely adopted, it relies on large-scale preference da...
By Yuxuan Zhang, Cong Wei, Penghui Du, Bo Li, Junwen Miao, Huaisong Zhang, Songcheng Cai, Yubo Wang, Dongfu Jiang, Yuyu Zhang, Changqian Yu, Ping Nie, Wenhu Chen, Kelsey R. Allen
arXiv:2610.16224v1 Announce Type: cross
Abstract: Text-to-image models have made remarkable progress in visual fidelity and instruction following, yet complex compositional generation remains challen...
By Zixu Wang, Xiaoran Zhao, Yahui Liu, Feng-ao Wang, Junhao Xiao, You Shu, Bing Liu, Yiting Liu, Songlei Jian, Zhiliang Tian, Dongsheng Li
arXiv:2606. 01213v1 Announce Type: cross Abstract: Despite tremendous recent progress, current text-guided image editing methods still struggle with many aspects of editing involving instruction following, minimally editing the source image, and ensuring high visual quality.
By Aishwarya Agrawal, Roy Hirsch, Yasumasa Onoe, Sherry Ben, Jason Baldridge
RubricRM introduces a pairwise generative reward modeling framework that generates an input‑specific rubric—comprising evaluation dimensions, weights, and scoring criteria—to score candidate images. The method is trained in two stages: supervised fine‑tuning to learn the rubric‑based scoring paradigm and GRPO to refine dimension‑level rewards. Experiments on text‑to‑image generation and instruction‑based image editing benchmarks demonstrate that RubricRM outperforms existing specialized reward models and competes with strong proprietary MLLM judges while using smaller backbones.
By Zijian Kan, Wei Wang, Long Luo, Bing Zhao, Xuan Ren, Weixu Qiao, Wenbo Li, Hu Wei, Lin Qu
arXiv:2610.19882v1 Announce Type: new
Abstract: Instruction-driven image editing has reached native 4K, yet how such editors are trained remains undisclosed, and pretraining one from scratch is beyon...
By Shanyan Guan, Shiyu Li, Xin Li, Zhizhou Chen, Yunzhe Xu, Yuehao Liu, Ruixue Shen, Shiju Zhang, Min Jin, Qiang Liu, Junbo Liu, Penggang Shi, Sheng Sun, Xiao Wang, Yong Zhang, Zhongang Qi, Wei Li, Chao Ma, Ying Tai, Yanhao Ge
VibeEdit is a new image‑editing tool that lets users draw spatial marks and add brief notes directly on an image to create a canvas instruction. These instructions guide the model to perform tasks such as adding, removing, replacing, or moving objects, as well as modifying attributes, without needing a separate text prompt. The system is trained on 1.55 million annotated edit pairs and uses a layer‑decoupled conditioning approach, achieving higher VLM rubric scores and PSNR on a benchmark that tests target selection among similar objects.
By Jinjing Zhao, Fangyun Wei, Yitong Wang, Xiuyu Wu, Yunuo Chen, Yang Yue, Sirui Zhang, Wenbo Wang, Hongyang Zhang, Dong Chen, Yan Lu, Chang Xu
arXiv:2609.01409v1 Announce Type: new
Abstract: Vision-language models (VLMs) have shown strong performance in generating scientific figures from text or images. However, producing publication-ready...
By Christian Greisinger, Zhixue Zhao, Steffen Eger
arXiv:2606. 27608v1 Announce Type: cross Abstract: We present Qwen-Image-2.
By Yixian Xu, Kaiyuan Gao, Yuxiang Chen, Yilei Chen, Zecheng Tang, Zihao Liu, Zikai Zhou, Deqing Li, Hao Meng, Kuan Cao, Jiahao Li, Jie Zhang, Liang Peng, Lihan Jiang, Ningyuan Tang, Shengming Yin, Tianhe Wu, Xiaoyue Chen, Yan Shu, Yanran Zhang, Yi Wang, Yu Wu, Yujia Wu, Zekai Zhang, Zhendong Wang, Xiao Xu, Kun Yan, Chenfei Wu
The paper introduces RC‑GRPO‑Editing, a region‑constrained Group Relative Policy Optimization framework for flow‑based image editing. It localizes exploration by decoupling initial noise perturbations to reduce background‑induced reward variance and adds an attention concentration reward to keep cross‑attention focused on the intended editing region. Experiments on CompBench demonstrate consistent gains in instruction adherence within the editing region while better preserving non‑target content.
By Zhuohan Ouyang, Zhe Qian, Wenhuo Cui, Chaoqun Wang
ACEdit introduces a training scheme that stabilizes multi-turn image editing by forcing the editor to operate on its own previous outputs, a concept called Self-Forcing. To overcome unreliable supervision from self-generated trajectories, the method employs a distillation objective using a teacher editor that remains anchored to the original image while receiving compound instructions. Experiments demonstrate reduced degradation over long editing sequences while preserving quality and instruction adherence.
By Sara Dorfman, Nir Goren, Gal Barak, Daniel Cohen-Or, Or Patashnik