Research map / Image generation / Editing and style transfer
OpenGPT-4o-Image: A Comprehensive Dataset for Advanced Image Generation and Editing
zhihong Chen, Xuehai Bai, Yang Shi, Chaoyou Fu, Huanyu Zhang, Haotian Wang, Xiaoyan Sun, Zhang Zhang, Liang Wang, Yuanxing Zhang, Pengfei Wan, YiFan Zhang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| fine-tune on OpenGPT-4o-Image 40K | MagicBrush · ImgEdit-Bench | overall score: 1.90 -> 2.30 | Improved |
| fine-tune on OpenGPT-4o-Image 40K | OmniGen2 · ImgEdit-Bench | overall score: 3.39 -> 3.82 | Improved |
| fine-tune on OpenGPT-4o-Image 40K | UniWorld-V1 · ImgEdit-Bench | overall score: 3.26 -> 3.86 | Improved |
| fine-tune on OpenGPT-4o-Image 40K | OmniGen · GEdit-Bench | average: 4.87 -> 5.55 | Improved |
| fine-tune on OpenGPT-4o-Image 40K | Harmon · GenEval | overall: 0.76 -> 0.86 | Improved |
| fine-tune on OpenGPT-4o-Image 40K | Harmon · DPG-Bench | overall: 81.27 -> 85.60 | Improved |
| fine-tune on OpenGPT-4o-Image 40K | OmniGen2 · GenEval | overall: 0.80 -> 0.82 | Improved |
| fine-tune on OpenGPT-4o-Image 40K | OmniGen2 · GEdit-Bench | average: 5.80 -> 6.31 | Improved |
| increase fine-tuning data from 20K to 40K (ablation) | UniWorld-V1 · GEdit + ImgEdit (Two Avg) | average: 6.46 -> 6.58 | Improved |
| increase fine-tuning data from 20K to 40K (ablation) | UniWorld-V1 · GEdit-Bench | average: 5.50 -> 5.79 -> 5.43 | Mixed |
| use OpenGPT-4o-Image instead of ShareGPT-4o-Image | UniWorld-V1 · ImgEdit-Bench | overall score: 3.7 -> 3.86 | Improved |
| use OpenGPT-4o-Image instead of ShareGPT-4o-Image | UniWorld-V1 · DPG-Bench | overall: 82.71 -> 83.66 | Improved |
Closest papers by meaning
- DuoGen: Towards Autonomous Interleaved Multimodal Generation (CVPR 2026 · similarity 0.88)
- OmniGen2: Towards Instruction-Aligned Multimodal Generation (CVPR 2026 · similarity 0.88)
- ScImage: How good are multimodal large language models at scientific text-to-image generation? (ICLR 2025 · similarity 0.87)
- ImgEdit: A Unified Image Editing Dataset and Benchmark (NeurIPS 2025 · similarity 0.87)
- Factuality Matters: When Image Generation and Editing Meet Structured Visuals (ICLR 2026 · similarity 0.87)
- MMIG-Bench: Towards Comprehensive and Explainable Evaluation of Multi-Modal Image Generation Models (NeurIPS 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
