Information diffusion prediction is a vital component for a wide range of social applications, including viral marketing identification and personal recommendation. Prior methods primarily focus on learning target user representation by modeling contextual information from the historical retweet user sequence of a single cascade, overlooking the uncertainties that exist in both historical propagation trajectory and future diffusion trends. In this work, we propose DucDiff, a novel dual-consistent diffusion model for enhancing target user representation used for information diffusion prediction. DucDiff harnesses the distribution generation capability of the diffusion model to generate target user representations from a distributional perspective rather than a fixed vector. Specifically, it captures the multi-latent aspects (i.e., uncertainties) of target user representation from historical and future user sequences, respectively, using disentangled dual denoising modules. Additionally, a shared information bottleneck is designed for the cross-distillation of knowledge between the historical and future denoising modules, eliminating the performance gap between training and inference, while ensuring that future information can be implicitly introduced during the inference phase. Extensive experiments conducted on five datasets demonstrate that DucDiff significantly outperforms state-of-the-art baselines.

Information diffusion prediction is a critical task for many social network applications. However, current methods are mainly limited by the following aspects: user relationships behind resharing behaviors are complex and entangled. To address these issues, we propose MHGFormer, a novel multi-channel hypergraph transformer framework, to better decouple complex user relations and obtain fine-grained user representations. First, we employ designed triangular motifs to decouple user relations into three different level hypergraphs. Second, a position-aware hypergraph transformer is used to refine user relation and obtain high-quality user representations. Extensive experiments conducted on two social datasets demonstrate that MHGFormer outperforms state-of-the-art diffusion models across several settings.

Accurately predicting information diffusion is critical for a vast range of applications. Existing methods generally consider user re-sharing behaviors to be driven by a single intent, and/or assume cascade temporal influence to be unchanged, which might not be consistent with real-world scenarios. To address these issues, we propose a self-supervised disentanglement framework (DisenIDP) for information diffusion prediction. First, we construct intent-aware hypergraphs to capture users' potential intents from different perspectives, and then perform the light hypergraph convolution to adaptively activate disentangled intents. Second, we extract long-term and short-term cascade influence via independent attention-based encoders. Finally, we set a self-supervised disentanglement task to alleviate the information loss and learn better-disentanglement representations. Extensive experiments conducted on two real-world social datasets demonstrate that DisenIDP outperforms state-of-the-art models across several settings.