Overcoming Barriers to Skill Injection in Language Modeling: Case Study in Arithmetic

TL;DR

This paper presents a novel framework with information-theoretic interventions to enable large language models to acquire mathematical reasoning skills without losing their linguistic capabilities.

Contribution

It introduces a new method to inject non-linguistic skills into language models while preventing catastrophic forgetting of linguistic knowledge.

Findings

Successful injection of arithmetic reasoning into language models

Retention of linguistic skills after skill injection

Enhanced performance on mathematical reasoning tasks

Abstract

Through their transfer learning abilities, highly-parameterized large pre-trained language models have dominated the NLP landscape for a multitude of downstream language tasks. Though linguistically proficient, the inability of these models to incorporate the learning of non-linguistic entities (numerals and arithmetic reasoning) limits their usage for tasks that require numeric comprehension or strict mathematical reasoning. However, as we illustrate in this paper, building a general purpose language model that also happens to be proficient in mathematical reasoning is not as straight-forward as training it on a numeric dataset. In this work, we develop a novel framework that enables language models to be mathematically proficient while retaining their linguistic prowess. Specifically, we offer information-theoretic interventions to overcome the catastrophic forgetting of linguistic skills that occurs while injecting non-linguistic skills into language models.