Blame for README.md

---
language:
- zh
- en
datasets:
- MMFact/MMFact
tags:
- mmfact
- multimodal
- fake-news-detection
---
<h1 align="center" style="text-align: center; font-size: 40px;">MMFact-CKPT</h1>

<p align="center">
  <img src="images/logo.png" alt="MMFact logo" width="280">
</p>

<p align="center" style="text-align: center;">
  <strong>Model Checkpoints for Multimodal Fake News Detection</strong>
</p>

<p align="center">
  <a href="https://arxiv.org/abs/2610.34162"><img src="https://img.shields.io/badge/arXiv-2610.34162-B31B1B?style=flat&amp;logo=arxiv&amp;logoColor=white" alt="arXiv: 2610.34162" height="20"></a>
  <a href="https://github.com/yuchenmiao04/MMFact"><img src="https://img.shields.io/badge/GitHub-MMFact-181717?style=flat&amp;logo=github&amp;logoColor=white" alt="GitHub: MMFact" height="20"></a>
  <a href="https://huggingface.co/datasets/MMFact/MMFact"><img src="https://img.shields.io/badge/Hugging%20Face-Dataset-D4A017?style=flat&amp;logo=huggingface&amp;logoColor=FFD21E" alt="Hugging Face: Dataset" height="20"></a>
</p>

**English** | [简体中文](https://huggingface.co/MMFact/MMFact-CKPT/blob/main/README_zh.md)

MMFact-CKPT is the model checkpoint repository for [MMFact](https://github.com/yuchenmiao04/MMFact), supporting reproducible experiments on Weibo-17, Weibo-21<sup>†</sup>, and GossipCop.

This repository contains checkpoints for **47 methods**, including **24 methods with separate ResNet and ViT variants**, covering three datasets and three random seeds for a total of **639 checkpoints**. Checkpoints are organized by dataset, method, variant, and seed to support evaluation and comparison using the MMFact toolbox.

## Checkpoint Coverage

| Method group | Methods | Variants | Checkpoints per method | Total |
|---|---:|---|---:|---:|
| Methods with a single variant | 23 | `default` | 9 | 207 |
| Methods with two visual variants | 24 | `resnet`, `vit` | 18 | 432 |
| **Total** | **47** | — | — | **639** |

Each variant has one checkpoint for each combination of the following datasets and random seeds:

| Setting | Values |
|---|---|
| Datasets | `weibo17`, `weibo21`, `gossipcop` |
| Random seeds | `0`, `42`, `2026` |
| Checkpoint filename | `checkpoint.pt` |

The `default` variant identifies methods without separate ResNet and ViT versions in this collection.

<sup>†</sup> Weibo-21 refers to the reconstructed version provided by MMFact, based on the [original paper](https://doi.org/10.1145/3459637.3482139), with a fixed 8:1:1 train/validation/test split. See the [dataset repository](https://huggingface.co/datasets/MMFact/MMFact) for details.

## Organization and Usage

All checkpoints follow a consistent directory structure:

```text
<dataset>/<method_slug>/<variant>/seed-<seed>/checkpoint.pt
```

For example, the EANN directory layout is as follows:

```text
weibo17/
└── eann/
    ├── resnet/
    │   ├── seed-0/
    │   │   └── checkpoint.pt
    │   ├── seed-42/
    │   │   └── checkpoint.pt
    │   └── seed-2026/
    │       └── checkpoint.pt
    └── vit/
        ├── seed-0/
        │   └── checkpoint.pt
        ├── seed-42/
        │   └── checkpoint.pt
        └── seed-2026/
            └── checkpoint.pt
```

Download an individual checkpoint using `huggingface_hub`:

```python
from huggingface_hub import hf_hub_download

checkpoint_path = hf_hub_download(
    repo_id="MMFact/MMFact-CKPT",
    repo_type="model",
    filename="weibo17/eann/resnet/seed-0/checkpoint.pt",
)

print(checkpoint_path)
```

Change the filename to select a checkpoint for another dataset, method, variant, or random seed.

Use the corresponding model implementation and configuration in the [MMFact toolbox](https://github.com/yuchenmiao04/MMFact) to load and evaluate the checkpoint. Dataset definitions, preprocessing resources, and precomputed features are documented in the [MMFact dataset repository](https://huggingface.co/datasets/MMFact/MMFact).

## Citation

If you use these checkpoints or other MMFact resources in your research, please cite our paper:

```bibtex
@misc{miao2026mmfactsimplifyingmultimodalmisinformation,
  title={MMFact: Simplifying Multimodal Misinformation Detection with Shared Resources and Tools},
  author={Yuchen Miao and Zijun Wang and Mingtai Zhang and Xiaoyi Chen and Sihan Deng and Xianglin Ding},
  year={2026},
  eprint={2610.34162},
  archivePrefix={arXiv},
  primaryClass={cs.MM},
  url={https://arxiv.org/abs/2610.34162},
}
```