Learn it or Leave it: Module Composition and Pruning for Continual Learning

TL;DR

This paper introduces MoCL-P, a lightweight continual learning method that combines module composition and adaptive pruning to prevent forgetting, transfer knowledge, and improve parameter efficiency across multiple tasks.

Contribution

MoCL-P is a novel approach that integrates task-guided module composition with adaptive pruning, avoiding parameter expansion in continual learning.

Findings

Achieves state-of-the-art performance on three benchmarks with up to 176 tasks.

Improves parameter efficiency by up to three times.

Effectively balances knowledge transfer and computational cost.

Abstract

In real-world environments, continual learning is essential for machine learning models, as they need to acquire new knowledge incrementally without forgetting what they have already learned. While pretrained language models have shown impressive capabilities on various static tasks, applying them to continual learning poses significant challenges, including avoiding catastrophic forgetting, facilitating knowledge transfer, and maintaining parameter efficiency. In this paper, we introduce MoCL-P, a novel lightweight continual learning method that addresses these challenges simultaneously. Unlike traditional approaches that continuously expand parameters for newly arriving tasks, MoCL-P integrates task representation-guided module composition with adaptive pruning, effectively balancing knowledge integration and computational overhead. Our evaluation across three continual learning benchmarks with up to 176 tasks shows that MoCL-P achieves state-of-the-art performance and improves parameter efficiency by up to three times, demonstrating its potential for practical applications where resource requirements are constrained.