KaLM at SemEval-2020 Task 4: Knowledge-aware Language Models for Comprehension And Generation

TL;DR

This paper introduces a knowledge-aware approach using large-scale pre-trained models for commonsense validation and explanation, demonstrating improved performance and achieving second place in subtask C of SemEval 2020.

Contribution

It proposes a novel evidence-searching method and utilizes multiple pre-trained models tailored for each subtask, advancing commonsense reasoning tasks.

Findings

Evidence-searching improves explanation accuracy.

Achieved 2nd place in subtask C based on human evaluation.

Model strategies enhance commonsense validation performance.

Abstract

This paper presents our strategies in SemEval 2020 Task 4: Commonsense Validation and Explanation. We propose a novel way to search for evidence and choose the different large-scale pre-trained models as the backbone for three subtasks. The results show that our evidence-searching approach improves model performance on commonsense explanation task. Our team ranks 2nd in subtask C according to human evaluation score.