Can LLMs Learn New Concepts Incrementally without Forgetting?

TL;DR

This paper investigates whether large language models can learn new concepts incrementally without forgetting, introduces a new dataset for evaluation, and analyzes factors affecting incremental learning performance.

Contribution

It introduces Concept-1K, a novel dataset for evaluating incremental learning in LLMs, and provides comprehensive analysis of factors influencing their ability to learn without forgetting.

Findings

LLMs still suffer from catastrophic forgetting.

LoRA fine-tuning may increase forgetting despite fewer parameter updates.

Model scale, buffer size, and pretraining impact IL performance.

Abstract

Large Language Models (LLMs) have achieved remarkable success across various tasks, yet their ability to learn incrementally without forgetting remains underexplored. Incremental learning (IL) is crucial as it enables models to acquire new knowledge while retaining previously learned information, akin to human learning. Existing benchmarks for IL are insufficient due to data leakage issues and the overqualification of LLMs. To address these challenges, we introduce Concept-1K, a novel dataset comprising 1,023 recently emerged concepts across diverse domains. The concepts in Concept-1K are discrete, interpretable units of knowledge that allow for fine-grained analysis of learning and forgetting processes. Using Concept-1K as a testbed, we aim to answer the question: ``Can LLMs learn new concepts incrementally without forgetting like humans?'' Our investigation reveals that LLMs still suffer from catastrophic forgetting and that LoRA, despite fine-tuning fewer parameters, may lead to more forgetting on training data. Additionally, we explore the roles of in-context learning, model scale, buffer size, and pretraining in IL performance. These findings highlight the strengths and limitations of LLMs in IL scenarios and provide a robust benchmark for future research.