Evaluating NLP Systems On a Novel Cloze Task: Judging the Plausibility of Possible Fillers in Instructional Texts

TL;DR

This paper introduces a novel NLP evaluation task that assesses the plausibility of filler words in cloze tasks, emphasizing absolute quality prediction across diverse inputs, and compares various models including ensemble approaches.

Contribution

It proposes a new plausibility classification task for cloze fillers, explores multiple architectures, and develops an ensemble method to enhance performance.

Findings

Ensemble methods outperform individual models.

The new task provides a more comprehensive evaluation of language understanding.

Detailed comparison of architectures informs future model development.

Abstract

Cloze task is a widely used task to evaluate an NLP system's language understanding ability. However, most of the existing cloze tasks only require NLP systems to give the relative best prediction for each input data sample, rather than the absolute quality of all possible predictions, in a consistent way across the input domain. Thus a new task is proposed: predicting if a filler word in a cloze task is a good, neutral, or bad candidate. Complicated versions can be extended to predicting more discrete classes or continuous scores. We focus on subtask A in Semeval 2022 task 7, explored some possible architectures to solve this new task, provided a detailed comparison of them, and proposed an ensemble method to improve traditional models in this new task.