GRADE: Benchmarking Discipline-Informed Reasoning in Image Editing
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908882176049152 |
|---|---|
| author | Liu, Mingxin Fan, Ziqian Wang, Zhaokai Gu, Leyao Zhu, Zirun He, Yiguo Yang, Yuchen Tian, Changyao Zhao, Xiangyu Liao, Ning Zhang, Shaofeng Ren, Qibing Zhong, Zhihang Zhou, Xuanhe Yan, Junchi Yang, Xue |
| author_facet | Liu, Mingxin Fan, Ziqian Wang, Zhaokai Gu, Leyao Zhu, Zirun He, Yiguo Yang, Yuchen Tian, Changyao Zhao, Xiangyu Liao, Ning Zhang, Shaofeng Ren, Qibing Zhong, Zhihang Zhou, Xuanhe Yan, Junchi Yang, Xue |
| contents | Unified multimodal models target joint understanding, reasoning, and generation, but current image editing benchmarks are largely confined to natural images and shallow commonsense reasoning, offering limited assessment of this capability under structured, domain-specific constraints. In this work, we introduce GRADE, the first benchmark to assess discipline-informed knowledge and reasoning in image editing. GRADE comprises 520 carefully curated samples across 10 academic domains, spanning from natural science to social science. To support rigorous evaluation, we propose a multi-dimensional evaluation protocol that jointly assesses Discipline Reasoning, Visual Consistency, and Logical Readability. Extensive experiments on 20 state-of-the-art open-source and closed-source models reveal substantial limitations in current models under implicit, knowledge-intensive editing settings, leading to large performance gaps. Beyond quantitative scores, we conduct rigorous analyses and ablations to expose model shortcomings and identify the constraints within disciplinary editing. Together, GRADE pinpoints key directions for the future development of unified multimodal models, advancing the research on discipline-informed image editing and reasoning. Our benchmark and evaluation code are publicly released. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_12264 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | GRADE: Benchmarking Discipline-Informed Reasoning in Image Editing Liu, Mingxin Fan, Ziqian Wang, Zhaokai Gu, Leyao Zhu, Zirun He, Yiguo Yang, Yuchen Tian, Changyao Zhao, Xiangyu Liao, Ning Zhang, Shaofeng Ren, Qibing Zhong, Zhihang Zhou, Xuanhe Yan, Junchi Yang, Xue Computer Vision and Pattern Recognition Unified multimodal models target joint understanding, reasoning, and generation, but current image editing benchmarks are largely confined to natural images and shallow commonsense reasoning, offering limited assessment of this capability under structured, domain-specific constraints. In this work, we introduce GRADE, the first benchmark to assess discipline-informed knowledge and reasoning in image editing. GRADE comprises 520 carefully curated samples across 10 academic domains, spanning from natural science to social science. To support rigorous evaluation, we propose a multi-dimensional evaluation protocol that jointly assesses Discipline Reasoning, Visual Consistency, and Logical Readability. Extensive experiments on 20 state-of-the-art open-source and closed-source models reveal substantial limitations in current models under implicit, knowledge-intensive editing settings, leading to large performance gaps. Beyond quantitative scores, we conduct rigorous analyses and ablations to expose model shortcomings and identify the constraints within disciplinary editing. Together, GRADE pinpoints key directions for the future development of unified multimodal models, advancing the research on discipline-informed image editing and reasoning. Our benchmark and evaluation code are publicly released. |
| title | GRADE: Benchmarking Discipline-Informed Reasoning in Image Editing |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.12264 |