cs.CVOct 30, 2026

From Evaluation to Evolution: Benchmarking Image Editing and Evolving Image Models

Authors: Jun He, Canmiao Fu, Ying Zhang, Zhipeng Huang, Chenyu Mu, Yazhen Xie, Chen Li, Jing Lyu, 

Organizations: Sun Yat-sen University · WeChat Vision, Tencent Inc. · Xidian University · Fudan University · 

Abstract

Real-world image editing requires coordinating multiple visual transformations with content preservation and reference constraints, yet aggregate scores can readily obscure unmet local requirements. Effective evaluation must therefore identify local failures and translate them into priorities for subsequent model improvement. We introduce WeGenEditBench, comprising 2,740 evaluation instances and 148 task types across single-image, text-intensive, and multi-reference editing. Pre-generated checklists shared across models decompose each request into atomic requirements; item-level deductions and visual evidence are established first and then summarized into scores for edit execution, preservation, visual quality, and task-specific correctness. To support the experience-acquisition component of recursive self-improvement (RSI), we develop WeGenEvoStudio, a unified multi-agent harness for RSI-oriented learning that integrates model-conditioned failure diagnosis, retrieval-grounded data construction, supervised adaptation, and re-evaluation. Among 15 evaluated configurations, GPT-Image-2 leads single-image editing with an Overall score of 4.076, while Seedream-5.0-Pro leads text-intensive editing at 4.291. Evolution experiments across four generation and editing benchmarks demonstrate substantial targeted gains: on their report-selected Top-5 single-image targets, FLUX.2 [dev] improves Overall from 1.083 to 3.517, while Qwen-Image-Edit-2511 gains 13.2%. Together, our benchmark and harness provide a foundation for developing more reliable image models through interpretable evaluation and feedback-driven learning.

Figures & tables

Appendix figures & tables67 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. GEditBench v2: A Human-Aligned Benchmark for General Image Editing

    Mar 30, 2026Zhangqi Jiang, Zheng Sun, Xianfang Zeng +7Image Editing EvaluationText-Guided Image Editing

  2. ReasonEdit: Towards Interpretable Image Editing Evaluation via Reinforcement Learning

    May 8, 2026Honghua Chen, Zitong Xu, Huiyu Duan +3Image Editing EvaluationText-Guided Image Editing