Multimodal information extraction on social media is a series of fundamental tasks to construct the multimodal knowledge graph. The tasks aim to extract the structural information in free texts with the incorporate images, including: multimodal named entity typing and multimodal relation extraction. However, the growing number of multimodal data implies a growing category set and the newly emerged entity types or relations should be recognized without additional training. To address the aforementioned challenges, we focus on the zero-shot multimodal information extraction tasks which require using textual and visual modalities for recognizing unseen categories. Compared with text-based zero-shot information extraction models, the existing multimodal ones make the textual and visual modalities aligned directly and exploit various fusion strategies to improve their performances. But the existing methods ignore the fine-grained semantic correlation of text-image pairs and samples. Therefore, we propose the multimodal graph-based variational mixture of experts network (MG-VMoE) which takes the MoE network as the backbone and exploits it for aligning multimodal representations in a fine-grained way. Considering to learn informative representations of multimodal data, we design each expert network as a variational information bottleneck to process two modalities in a uni-backbone. Moreover, we also propose the multimodal graph-based virtual adversarial training to learn the semantic correlation between the samples. The experimental results on the two benchmark datasets demonstrate the superiority of MG-VMoE over the baselines.

翻译：社交媒体上的多模态信息抽取是构建多模态知识图谱的一系列基础任务。这些任务旨在结合图像从自由文本中提取结构化信息，包括：多模态命名实体分类和多模态关系抽取。然而，日益增长的多模态数据意味着类别集合的不断扩大，新出现的实体类型或关系应在无需额外训练的情况下被识别。为应对上述挑战，我们聚焦于零样本多模态信息抽取任务，该任务要求利用文本和视觉模态来识别未见过的类别。与基于文本的零样本信息抽取模型相比，现有的多模态模型直接将文本与视觉模态对齐，并采用多种融合策略以提升性能。但现有方法忽略了文本-图像对及样本间的细粒度语义关联。为此，我们提出了基于多模态图结构的变分专家混合网络（MG-VMoE），该网络以专家混合网络为骨干，并以细粒度方式对齐多模态表征。为学习多模态数据的有效表征，我们将每个专家网络设计为变分信息瓶颈，在统一骨干网络中处理两种模态。此外，我们还提出了基于多模态图的虚拟对抗训练方法，以学习样本间的语义关联。在两个基准数据集上的实验结果证明了MG-VMoE相较于基线模型的优越性。