README.md

---
base_model: facebook/dinov3-vitl16-pretrain-lvd1689m
datasets:
- DocTamperV1
- vankey/RealText-V2
- Jason37437/RealText-V2-Syn25k
language: en
library_name: pytorch
license: mit
metrics:
- f1
pipeline_tag: image-segmentation
tags:
- document-forgery-detection
- tampering-detection
- image-manipulation
- vision-transformer
- lora
---

# SEED Detector

This repository contains the official detector model for **SEED**, presented in the paper [SEED: Simple ViT and Evolving Harness for Explainable Text Forgery Detection](https://huggingface.co/papers/2606.21138).

**SEED Detector** is a lightweight vision transformer model for document forgery detection. It localizes tampered regions in document images and classifies images as real or forged.

## Architecture

| Component | Detail |
|-----------|--------|
| Backbone | [DINOv3 ViT-L/16](https://huggingface.co/facebook/dinov3-vitl16-pretrain-lvd1689m) |
| Finetuning | LoRA (rank=1, attention + MLP) |
| Queries | 1 mask query |
| Decoder blocks | 4 |
| Input size | 512 × 512 |
| Parameters | ~304M (only ~1M trainable with LoRA) |

## Usage

**Repository:** [GitHub](https://github.com/KahimWong/GenText-Forensics-3rd-Place)  |  **Checkpoint:** [Jason37437/SEED](https://huggingface.co/Jason37437/SEED) / [Google Drive](https://drive.google.com/file/d/1XRbcE2eEdSBdQbyiImg5w9Dn5oMRMKhv/view?usp=drive_link)

```python
from model.hf_wrapper import EoMTForTamperingDetection

model = EoMTForTamperingDetection.from_pretrained("Jason37437/SEED")
model.eval()

# The model outputs:
#   - mask_logits: per-query segmentation masks
#   - class_logits: per-query foreground/background scores  
#   - image_logits: image-level real vs forged classification
```

## Performance

### Localization (pixel-level F1)
| Dataset | F1 |
|---------|-----|
| T-SROIE | 0.782 |
| OSTF | 0.718 |
| TPIC-13 | 0.798 |
| RTM | 0.178 |
| Avg | 0.619 |

### Detection (image-level F1)
| Dataset | F1 |
|---------|-----|
| T-SROIE | 0.738 |
| OSTF | 0.832 |
| TPIC-13 | 0.930 |
| RTM | 0.207 |
| Avg | 0.677 |

## Citation

```bibtex
@article{wong2026seed,
  title={SEED: Simple ViT and Evolving Harness for Explainable Text Forgery Detection},
  author={Wong, Kahim and others},
  journal={arXiv preprint arXiv:2606.21138},
  year={2026}
}
```

## License

MIT License.