Research map / Image generation / Editing and style transfer
An Extensive Benchmark for Single-Round and Multi-Round Instruction-Based Image Editing
Yiwei Ma, Ke Ye, Weihuang Lin, Jiayi Ji, Xiaoshuai Sun, Tat-Seng Chua, Rongrong Ji
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 |
|---|---|---|---|
| evaluate in multi-round instead of single-round editing (Direction Perception) | Qwen-image · I2EBench2.0 | score: 83.09 -> 57.23 | Hurt |
| evaluate in multi-round instead of single-round editing (Counting) | Qwen-image · I2EBench2.0 | score: 65.19 -> 11.52 | Hurt |
| evaluate in multi-round instead of single-round editing (Counting) | FLUX Kontext · I2EBench2.0 | score: 49.63 -> 2.42 | Hurt |
| evaluate in multi-round instead of single-round editing (Object Removal) | Qwen-image · I2EBench2.0 | score: 78.42 -> 43.20 | Hurt |
| use diverse ChatGPT-rewritten instructions instead of original instructions (Counting) | Qwen-image · I2EBench2.0 | score: 65.19 -> 59.26 | Hurt |
| use diverse instructions instead of original (Counting) | FLUX Kontext · I2EBench2.0 | score: 49.63 -> 45.19 | Hurt |
| use diverse instructions instead of original (Counting) | HIVE · I2EBench2.0 | score: 18.57 -> 13.57 | Hurt |
| use Qwen3VL-8B as judge instead of the default judge (Counting) | Qwen-image · I2EBench2.0 | score: 65.19 -> 66.67 | No clear effect |
| use Qwen3VL-8B as judge instead of the default judge (Counting) | FLUX Kontext · I2EBench2.0 | score: 49.63 -> 50.37 | No clear effect |
| use FLUX Kontext instead of Qwen-image for Watermark Removal | FLUX Kontext · I2EBench2.0 | score: 47.10 -> 92.24 | Improved |
Closest papers by meaning
- I2EBench: A Comprehensive Benchmark for Instruction-based Image Editing (NeurIPS 2024 · similarity 0.95)
- Balancing Preservation and Modification: A Region and Semantic Aware Metric for Instruction-Based Image Editing (ICML 2025 · similarity 0.93)
- EdiVal-Agent: An Object-Centric Framework for Automated, Fine-Grained Evaluation of Multi-Turn Editing (ICLR 2026 · similarity 0.91)
- InsightEdit: Towards Better Instruction Following for Image Editing (CVPR 2025 · similarity 0.91)
- ImgEdit: A Unified Image Editing Dataset and Benchmark (NeurIPS 2025 · similarity 0.90)
- SmartEdit: Exploring Complex Instruction-based Image Editing with Multimodal Large Language Models (CVPR 2024 · similarity 0.89)
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.
