Bridge to Target Domain by Prototypical Contrastive Learning and Label Confusion: Re-explore Zero-Shot Learning for Slot Filling

TL;DR

This paper introduces a novel zero-shot slot filling method using prototypical contrastive learning and dynamic label confusion, significantly improving unseen slot performance and setting new state-of-the-art results.

Contribution

The paper proposes a new approach combining prototypical contrastive learning with label confusion to enhance knowledge transfer in zero-shot slot filling.

Findings

Significant improvement on unseen slots.

Achieved new state-of-the-art performance.

Effective knowledge transfer across domains.

Abstract

Zero-shot cross-domain slot filling alleviates the data dependence in the case of data scarcity in the target domain, which has aroused extensive research. However, as most of the existing methods do not achieve effective knowledge transfer to the target domain, they just fit the distribution of the seen slot and show poor performance on unseen slot in the target domain. To solve this, we propose a novel approach based on prototypical contrastive learning with a dynamic label confusion strategy for zero-shot slot filling. The prototypical contrastive learning aims to reconstruct the semantic constraints of labels, and we introduce the label confusion strategy to establish the label dependence between the source domains and the target domain on-the-fly. Experimental results show that our model achieves significant improvement on the unseen slots, while also set new state-of-the-arts on slot filling task.