QLSC: A Query Latent Semantic Calibrator for Robust Extractive Question Answering

TL;DR

The paper introduces QLSC, a novel auxiliary module for MRC models that enhances robustness to format variations in queries by capturing latent semantic features, improving answer accuracy.

Contribution

It proposes a new semantic calibrator with a scaling strategy and attention mechanism to deepen query-passage understanding in extractive QA models.

Findings

Improves robustness to format-variant queries

Enhances accuracy in pinpointing correct answers

Effective on robust Question-Answer datasets

Abstract

Extractive Question Answering (EQA) in Machine Reading Comprehension (MRC) often faces the challenge of dealing with semantically identical but format-variant inputs. Our work introduces a novel approach, called the ``Query Latent Semantic Calibrator (QLSC)'', designed as an auxiliary module for existing MRC models. We propose a unique scaling strategy to capture latent semantic center features of queries. These features are then seamlessly integrated into traditional query and passage embeddings using an attention mechanism. By deepening the comprehension of the semantic queries-passage relationship, our approach diminishes sensitivity to variations in text format and boosts the model's capability in pinpointing accurate answers. Experimental results on robust Question-Answer datasets confirm that our approach effectively handles format-variant but semantically identical queries, highlighting the effectiveness and adaptability of our proposed method.