Dual Path Modeling for Semantic Matching by Perceiving Subtle Conflicts

TL;DR

This paper introduces a dual path modeling framework that enhances semantic matching models' ability to detect subtle differences in sentence pairs by separately modeling affinity and difference semantics, leading to improved performance.

Contribution

The paper proposes a novel Dual Path Modeling Framework and DPM-Net to better perceive subtle semantic differences in sentence pairs, addressing limitations of existing transformer-based models.

Findings

Achieves consistent improvements over baselines on 10 datasets.

Effectively captures subtle semantic differences.

Enhances robustness in semantic matching tasks.

Abstract

Transformer-based pre-trained models have achieved great improvements in semantic matching. However, existing models still suffer from insufficient ability to capture subtle differences. The modification, addition and deletion of words in sentence pairs may make it difficult for the model to predict their relationship. To alleviate this problem, we propose a novel Dual Path Modeling Framework to enhance the model's ability to perceive subtle differences in sentence pairs by separately modeling affinity and difference semantics. Based on dual-path modeling framework we design the Dual Path Modeling Network (DPM-Net) to recognize semantic relations. And we conduct extensive experiments on 10 well-studied semantic matching and robustness test datasets, and the experimental results show that our proposed method achieves consistent improvements over baselines.