README.md

---
license: mit
task_categories:
- image-text-to-text
- text-to-image
tags:
- text-to-image
- evaluation
- artifacts
---

# MagicData340K: A Large-Scale Dataset for Fine-Grained Artifacts Assessment in Text-to-Image Generation

This repository hosts **MagicData340K**, a large-scale human-annotated dataset central to the [MagicMirror framework](https://wj-inf.github.io/MagicMirror-page). The MagicMirror framework introduces a comprehensive approach for the systematic and fine-grained evaluation of physical artifacts (such as anatomical and structural flaws) in Text-to-Image (T2I) generation.

`MagicData340K` is the first human-annotated large-scale dataset, comprising 340,000 generated images, each with fine-grained artifact labels. These annotations are guided by a detailed taxonomy of generated image artifacts, making the dataset crucial for understanding and improving the perceptual quality of T2I models.

**Paper**: [MagicMirror: A Large-Scale Dataset and Benchmark for Fine-Grained Artifacts Assessment in Text-to-Image Generation](https://arxiv.org/abs/2509.10260)

**Project Page**: [https://wj-inf.github.io/MagicMirror-page/](https://wj-inf.github.io/MagicMirror-page)

**Code (MagicMirror Benchmark)**: [https://github.com/wj-inf/MagicMirror](https://github.com/wj-inf/MagicMirror)

<p align="center"><img src="https://github.com/wj-inf/MagicMirror/blob/main/assets/output_example.png?raw=true" width="95%"></p>

## Related Hugging Face Assets

*   **Dataset (Self-reference)**: [wj-inf/MagicData340k](https://huggingface.co/datasets/wj-inf/MagicData340k)
*   **Model (MagicAssessor VLM)**: [wj-inf/MagicAssessor-7B](https://huggingface.co/wj-inf/MagicAssessor-7B)

## Sample Usage

The MagicMirror framework, which utilizes this dataset, allows for the assessment of Text-to-Image (T2I) models. After setting up the environment as detailed in the [MagicMirror GitHub repository](https://github.com/wj-inf/MagicMirror), you can organize your image data (e.g., as `./output/sdxl/merged_result_sdxl.jsonl`) and run the assessment script:

```bash
bash run.sh flux-schnell sdxl
```

## Citation

If you find MagicData340K or the MagicMirror framework useful for your research, please cite the paper:

```bibtex
@article{wang2025magicmirror,
  title   = {MagicMirror: A Large-Scale Dataset and Benchmark for Fine-Grained Artifacts Assessment in Text-to-Image Generation},
  author  = {Wang, Jia and Hu, Jie and Ma, Xiaoqi and Ma, Hanghang and Zeng, Yanbing and Wei, Xiaoming},
  journal = {arXiv preprint arXiv:2509.10260},
  year    = {2025}
}
```