Official-NV: An LLM-Generated News Video Dataset for Multimodal Fake News Detection
This addresses multimodal fake news detection for media and AI researchers, but it is incremental as it focuses on dataset creation and a baseline model.
The authors tackled the problem of multimodal fake news detection by constructing a new dataset called Official-NV from officially published news videos, augmented with LLM generation and manual verification, and proposed a baseline model OFNVD that uses GLU attention and cross-modal Transformers, showing effectiveness in benchmarking.
News media, especially video news media, have penetrated into every aspect of daily life, which also brings the risk of fake news. Therefore, multimodal fake news detection has recently garnered increased attention. However, the existing datasets are comprised of user-uploaded videos and contain an excess amounts of superfluous data, which introduces noise into the model training process. To address this issue, we construct a dataset named Official-NV, comprising officially published news videos. The crawl officially published videos are augmented through the use of LLMs-based generation and manual verification, thereby expanding the dataset. We also propose a new baseline model called OFNVD, which captures key information from multimodal features through a GLU attention mechanism and performs feature enhancement and modal aggregation via a cross-modal Transformer. Benchmarking the dataset and baselines demonstrates the effectiveness of our model in multimodal news detection.