Prototype-Regularized Federated Learning for Cross-Domain Aspect Sentiment Triplet Extraction

TL;DR

This paper introduces a federated learning framework using prototypes for cross-domain aspect sentiment triplet extraction, addressing data privacy and domain heterogeneity.

Contribution

It proposes a novel prototype-regularized federated learning method with a weighted aggregation and contrastive regularization for ASTE.

Findings

Outperforms baseline methods on four ASTE datasets.

Reduces communication costs compared to traditional federated learning.

Effectively captures shared features across domains despite data heterogeneity.

Abstract

Aspect Sentiment Triplet Extraction (ASTE) aims to extract all sentiment triplets of aspect terms, opinion terms, and sentiment polarities from a sentence. Existing methods are typically trained on individual datasets in isolation, failing to jointly capture the common feature representations shared across domains. Moreover, data privacy constraints prevent centralized data aggregation. To address these challenges, we propose Prototype-based Cross-Domain Span Prototype extraction (PCD-SpanProto), a prototype-regularized federated learning framework to enable distributed clients to exchange class-level prototypes instead of full model parameters. Specifically, we design a weighted performance-aware aggregation strategy and a contrastive regularization module to improve the global prototype under domain heterogeneity and the promotion between intra-class compactness and inter-class separability across clients. Extensive experiments on four ASTE datasets demonstrate that our method outperforms baselines and reduces communication costs, validating the effectiveness of prototype-based cross-domain knowledge transfer.