Visual Pivoting for (Unsupervised) Entity Alignment
This provides an unsupervised solution for entity alignment in knowledge graphs, addressing limitations in availability for tasks like cross-lingual alignment, though it is incremental in leveraging existing visual data.
The paper tackles the problem of aligning entities across heterogeneous knowledge graphs without human-labeled data by using visual semantic representations, achieving state-of-the-art performance on benchmark datasets like DBP15k and DWY15k, with images proving especially useful for aligning long-tail entities.
This work studies the use of visual semantic representations to align entities in heterogeneous knowledge graphs (KGs). Images are natural components of many existing KGs. By combining visual knowledge with other auxiliary information, we show that the proposed new approach, EVA, creates a holistic entity representation that provides strong signals for cross-graph entity alignment. Besides, previous entity alignment methods require human labelled seed alignment, restricting availability. EVA provides a completely unsupervised solution by leveraging the visual similarity of entities to create an initial seed dictionary (visual pivots). Experiments on benchmark data sets DBP15k and DWY15k show that EVA offers state-of-the-art performance on both monolingual and cross-lingual entity alignment tasks. Furthermore, we discover that images are particularly useful to align long-tail KG entities, which inherently lack the structural contexts necessary for capturing the correspondences.