Knowledge-Aware Language Model Pretraining

TL;DR

This paper introduces a simple method to enhance pretrained language models with explicit knowledge signals by signaling entities during pretraining, leading to improved factual accuracy and downstream task performance without altering the model architecture.

Contribution

The authors propose a knowledge-aware pretraining approach that signals entities via an extended tokenizer and an additional prediction task, improving knowledge encoding without architectural changes.

Findings

Enhanced factual correctness in knowledge probing tasks

Improved zero-shot question-answering performance

More semantically rich hidden representations

Abstract

How much knowledge do pretrained language models hold? Recent research observed that pretrained transformers are adept at modeling semantics but it is unclear to what degree they grasp human knowledge, or how to ensure they do so. In this paper we incorporate knowledge-awareness in language model pretraining without changing the transformer architecture, inserting explicit knowledge layers, or adding external storage of semantic information. Rather, we simply signal the existence of entities to the input of the transformer in pretraining, with an entity-extended tokenizer; and at the output, with an additional entity prediction task. Our experiments show that solely by adding these entity signals in pretraining, significantly more knowledge is packed into the transformer parameters: we observe improved language modeling accuracy, factual correctness in LAMA knowledge probing tasks, and semantics in the hidden representations through edge probing.We also show that our knowledge-aware language model (KALM) can serve as a drop-in replacement for GPT-2 models, significantly improving downstream tasks like zero-shot question-answering with no task-related training.