Knowledge Rumination for Pre-trained Language Models

TL;DR

This paper introduces Knowledge Rumination, a simple prompt-based method that helps pre-trained language models better utilize their internal knowledge for knowledge-intensive NLP tasks, improving performance without external retrieval.

Contribution

It proposes a novel prompt-based paradigm enabling PLMs to review and consolidate their latent knowledge, enhancing their effectiveness on various NLP benchmarks.

Findings

Improved performance on six commonsense reasoning tasks

Enhanced results on GLUE benchmarks

Effective across multiple language models

Abstract

Previous studies have revealed that vanilla pre-trained language models (PLMs) lack the capacity to handle knowledge-intensive NLP tasks alone; thus, several works have attempted to integrate external knowledge into PLMs. However, despite the promising outcome, we empirically observe that PLMs may have already encoded rich knowledge in their pre-trained parameters but fail to fully utilize them when applying them to knowledge-intensive tasks. In this paper, we propose a new paradigm dubbed Knowledge Rumination to help the pre-trained language model utilize that related latent knowledge without retrieving it from the external corpus. By simply adding a prompt like "As far as I know" to the PLMs, we try to review related latent knowledge and inject them back into the model for knowledge consolidation. We apply the proposed knowledge rumination to various language models, including RoBERTa, DeBERTa, and GPT-3. Experimental results on six commonsense reasoning tasks and GLUE benchmarks demonstrate the effectiveness of our proposed approach, which proves that the knowledge stored in PLMs can be better exploited to enhance performance. Code is available in https://github.com/zjunlp/knowledge-rumination.