PseudoReasoner: Leveraging Pseudo Labels for Commonsense Knowledge Base Population

TL;DR

PseudoReasoner is a semi-supervised framework that leverages pseudo labels generated by a teacher model to improve commonsense knowledge base population, especially enhancing out-of-domain generalization.

Contribution

It introduces a novel semi-supervised learning approach using pseudo labels and a filtering procedure, achieving state-of-the-art results in CSKB population.

Findings

Improves KG-BERT performance by 3.3 points overall.

Enhances out-of-domain performance by 5.3 points.

Achieves state-of-the-art results on CSKB population.

Abstract

Commonsense Knowledge Base (CSKB) Population aims at reasoning over unseen entities and assertions on CSKBs, and is an important yet hard commonsense reasoning task. One challenge is that it requires out-of-domain generalization ability as the source CSKB for training is of a relatively smaller scale (1M) while the whole candidate space for population is way larger (200M). We propose PseudoReasoner, a semi-supervised learning framework for CSKB population that uses a teacher model pre-trained on CSKBs to provide pseudo labels on the unlabeled candidate dataset for a student model to learn from. The teacher can be a generative model rather than restricted to discriminative models as previous works. In addition, we design a new filtering procedure for pseudo labels based on influence function and the student model's prediction to further improve the performance. The framework can improve the backbone model KG-BERT (RoBERTa-large) by 3.3 points on the overall performance and especially, 5.3 points on the out-of-domain performance, and achieves the state-of-the-art. Codes and data are available at https://github.com/HKUST-KnowComp/PseudoReasoner.