Event Process Typing via Hierarchical Optimal Transport

Abstract

Understanding intention behind event processes in texts is important to many applications. One challenging task in this line is event process typing, which aims to tag the process with one action label and one object label describing the overall action of the process and object the process likely affects respectively. To tackle this task, existing methods mainly rely on the matching of the event process level and label level representation, which ignores two important characteristics: Process Hierarchy and Label Hierarchy. In this paper, we propose a Hierarchical Optimal Transport (HOT) method to address the above problem. Specifically, we first explicitly extract the process hierarchy and label hierarchy. Then the HOT optimally matches the two types of hierarchy. Experimental results show that our model outperforms the baseline models, illustrating the effectiveness of our model.

Multi-Prototype Space Learning for Commonsense-Based Scene Graph Generation

“…TOT achieves a trade-off between predicate-wise comparison and triple-wise comparison when computing the optimal transport matrix, in which the significance of the two terms is controlled by the hyperparameter β ∈ [0, 1]. For the marginals of the transport matrix, instead of imposing strict equality constraints (Zhou et al 2023), we add two regularizers to penalize the KL-divergences between them and uniform distributions (Wu et al 2023). The terms 1 I 1 I and 1 J 1 J in Eqn.…”