Learn or Recall? Revisiting Incremental Learning with Pre-trained Language Models

2023/12/13 by Junhao Zheng, Zheng, Junhao, Shengjie Qiu +3 · 5 citations
Computer Science · #Domain Adaptation and Few-Shot Learning #Multimodal Machine Learning Applications #Topic Modeling

paper · pdf · doi:10.48550/arxiv.2312.07887

Abstract

Incremental Learning (IL) has been a long-standing problem in both vision and Natural Language Processing (NLP) communities. In recent years, as Pre-trained Language Models (PLMs) have achieved remarkable progress in various NLP downstream tasks, utilizing PLMs as backbones has become a common practice in recent research of IL in NLP. Most assume that catastrophic forgetting is the biggest obstacle to achieving superior IL performance and propose various techniques to overcome this issue. However, we find that this assumption is problematic. Specifically, we revisit more than 20 methods on four classification tasks (Text Classification, Intent Classification, Relation Extraction, and Named Entity Recognition) under the two most popular IL settings (Class-Incremental and Task-Incremental) and reveal that most of them severely underestimate the inherent anti-forgetting ability of PLMs. Based on the observation, we propose a frustratingly easy method called SEQ* for IL with PLMs. The results show that SEQ* has competitive or superior performance compared to state-of-the-art (SOTA) IL methods and requires considerably less trainable parameters and training time. These findings urge us to revisit the IL with PLMs and encourage future studies to have a fundamental understanding of the catastrophic forgetting in PLMs. The data, code and scripts are publicly available at https://github.com/zzz47zzz/codebase-for-incremental-learning-with-llm.

Cited by

• Mitigating Forgetting in Low Rank Adaptation
• DRL: Discriminative Representation Learning with Parallel Adapters for Class Incremental Learning
• MCITlib: Multimodal Continual Instruction Tuning Library and Benchmark
• Revisiting Continual Semantic Segmentation with Pre-trained Vision Models
• MLLM-CTBench: A Benchmark for Continual Instruction Tuning with Reasoning Process Diagnosis

Related

• Can LLMs Learn New Concepts Incrementally without Forgetting?
• Catastrophic Forgetting in LLMs: A Comparative Analysis Across Language Tasks
• Unlearning That Lasts: Utility-Preserving, Robust, and Almost Irreversible Forgetting in LLMs
• Examining Forgetting in Continual Pre-training of Aligned Large Language Models
• Towards Incremental Learning in Large Language Models: A Critical Review