DuQM: A Chinese Dataset of Linguistically Perturbed Natural Questions for Evaluating the Robustness of Question Matching Models

TL;DR

This paper introduces DuQM, a comprehensive Chinese dataset with linguistically perturbed questions to evaluate and diagnose the robustness of question matching models, revealing limitations of artificial adversarial examples.

Contribution

It presents DuQM, a novel dataset with diverse linguistic perturbations for Chinese question matching robustness evaluation, enabling detailed model diagnosis.

Findings

DuQM effectively distinguishes different question matching models.

Linguistic perturbations reveal model strengths and weaknesses.

Artificial adversarial examples are less effective on natural texts.

Abstract

In this paper, we focus on studying robustness evaluation of Chinese question matching. Most of the previous work on analyzing robustness issue focus on just one or a few types of artificial adversarial examples. Instead, we argue that it is necessary to formulate a comprehensive evaluation about the linguistic capabilities of models on natural texts. For this purpose, we create a Chinese dataset namely DuQM which contains natural questions with linguistic perturbations to evaluate the robustness of question matching models. DuQM contains 3 categories and 13 subcategories with 32 linguistic perturbations. The extensive experiments demonstrate that DuQM has a better ability to distinguish different models. Importantly, the detailed breakdown of evaluation by linguistic phenomenon in DuQM helps us easily diagnose the strength and weakness of different models. Additionally, our experiment results show that the effect of artificial adversarial examples does not work on the natural texts.