Official repository for "FakeSV: A Multimodal Benchmark with Rich Social Context for Fake News Detection on Short Video Platforms", AAAI 2023. (Please note that the arxiv version is more complete.)
- Dataset: The video ID (which can be used to infer the video URL) and corresponding annotations have been released. Also, we provide two data split used in the paper, i.e. event-based and temporal.
- Models: We reproduce some SOTA methods on fake news video detection to provide benchmark results for FakeSV. Codes for our proposed model SV-FEND and other methods are provided.
Anaconda 4.13.0, python 3.8.5, pytorch 1.10.1 and cuda 11.7. For other libs, please refer to the file requirements.txt.
Please sign this agreement and send the signed copy through your institutional email to [email protected].
You could use the above repositories to extract features by yourself, or use our pre-extracted features (VGG19/C3D/VGGish). Besides, we also provide five-fold checkpoints for comparison.
@inproceedings{fakesv,
title={FakeSV: A Multimodal Benchmark with Rich Social Context for Fake News Detection on Short Video Platforms},
author={Qi, Peng and Bu, Yuyan and Cao, Juan and Ji, Wei and Shui, Ruihao and Xiao, Junbin and Wang, Danding and Chua, Tat-Seng},
booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
year={2023},
organization={AAAI}
}
Related Survey (ACM Multimedia 2023) and Companion GitHub Repository :
@inproceedings{mvdsurvey,
title={Combating Online Misinformation Videos: Characterization, Detection, and Future Directions},
author={Bu, Yuyan and Sheng, Qiang and Cao, Juan and Qi, Peng and Wang, Danding and Li, Jintao},
booktitle={Proceedings of the 31st ACM International Conference on Multimedia},
year={2023},
doi={10.1145/3581783.3612426},
publisher = {Association for Computing Machinery},
}
Related Method Paper (ACL 2023 Findings)
@inproceedings{mvdsurvey,
title={Two Heads Are Better Than One: Improving Fake News Video Detection by Correlating with Neighbors},
author={Qi, Peng and Zhao, Yuyang and Shen, Yufeng and Ji, Wei and Cao, Juan and Chua, Tat-Seng},
booktitle={Findings of the Association for Computational Linguistics: ACL 2023},
year={2023},
doi={10.18653/v1/2023.findings-acl.756},
pages = "11947--11959",
publisher = {Association for Computational Linguistics},
}
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent