IQA-T1

[ECCV 2026] IQA-T1: Tool-based Visual Evidence Reasoning for Image Quality Assessment

This repository provides the pretrained weights of IQA-T1 built on the Qwen3-VL-4B-Instruct backbone.

Quick Links

• 📄 Paper
• 💻 GitHub
• 🎮 Online Demo
• 📊 Q-Tool Dataset

Usage

See the GitHub repository for training and inference code, or try the online demo directly in your browser.

Citation

@inproceedings{wu2026iqat1,
  title={IQA-T1: Tool-based Visual Evidence Reasoning for Image Quality Assessment},
  author={Wu, Jinjian and Tang, Jiaqi and Wei, Wei and Yan, Yingying and Chen, Jianmin and Geng, Botong and Zhang, Lei and Chen, Qifeng},
  booktitle={European Conference on Computer Vision (ECCV)},
  year={2026}
}