Let the Bullets Fly: Multimodal Fake News Detection with Temporal-Aligned Generative Danmaku

The social interactions among crowds via $\mathit{\text{Danmaku}}$ (a.k.a., bullet comments) on modern multimedia platforms can facilitate both viewpoint conflicts and consensus, providing fine-grained discriminative social signals that can benefit fake news detection. However, the inherent accumulation latency of $\mathit{\text{Danmaku}}$ in real-world scenarios violates the real-time necessity of fake news detection, making the studies of $\mathit{\text{Danmaku}}$-related fake news detection underexplored. To break this violation, we simulate this temporal-aware user interactive process by proposing a novel temporal $\mathbf{\text{Gen}}$erative $\mathbf{\text{da}}$nmaku framework, called $\mathbf{\text{Genda}}$, which consists of: (1) a $\mathit{\text{Danmaku}}$ Trigger for predicting the timing and intensity of user reactions; and (2) a $\mathit{\text{Danmaku}}$ Generator for synthesizing corresponding semantic and emotional expressions, thereby mutually constructing a temporally aligned and human-like pseudo $\mathit{\text{Danmaku}}$ streams. To make the generated $\mathit{\text{Danmaku}}$ useful for identifying fake news videos, we further design a $\mathit{\text{Danmaku}}$-guided Temporal Multimodal fake news detection model - $\mathbf{\text{DM-FEND}}$, which enables fine-grained multimodal interactions among video, audio, text, and $\mathit{\text{Danmaku}}$, enhancing dynamic modalities alignment and semantic noise inhibition. The experimental results demonstrate that \emph{DM-FEND} consistently outperforms state-of-the-art baselines across both Chinese (FakeSV) and English (FakeTT) benchmarks. Further ablations validate the crucial role of temporal $\mathit{\text{Danmaku}}$ modeling in enhancing robustness and discriminative capability. Finally, this study offers a bright and robust solution for multimodal fake news detection in modern social interactive fashions by bridging the temporal inconsistency between news and user behaviors.