用信息量筛选参数，让预训练模型持续学习更灵活高效
Sparse Tuning Enhances Plasticity in PTM-based Continual Learning
- 基于互信息敏感度，只更新少于5%的关键参数
- 每步更新参数少于0.5%，显著减少对预训练知识的干扰
- 无需改动原有框架，可提升多种持续学习方法性能
基于预训练模型的持续学习在顺序任务中具有高效适应潜力。然而，现有方法多冻结预训练模型并依赖提示或适配器等辅助模块，限制了模型可塑性，在分布显著偏移时导致泛化不佳。全量微调虽能提升适应能力，但可能破坏关键预训练知识。本文提出互信息引导的稀疏微调（MIST），一种即插即用的方法，依据对互信息目标的敏感度，仅选择性更新少于5%的模型参数。MIST在保持泛化能力的同时实现有效的任务特定适配。为进一步降低干扰，我们引入强稀疏正则化：在微调过程中随机丢弃梯度，使每步更新参数少于0.5%。该方法在标准冻结类方法前应用，可在多个持续学习基准上一致提升性能。实验表明，将其集成至多个基线均带来显著性能增益。代码已开源：https://github.com/zhwhu/MIST。
原文摘要 · Abstract (English)
Continual Learning with Pre-trained Models holds great promise for efficient adaptation across sequential tasks. However, most existing approaches freeze PTMs and rely on auxiliary modules like prompts or adapters, limiting model plasticity and leading to suboptimal generalization when facing significant distribution shifts. While full fine-tuning can improve adaptability, it risks disrupting crucial pre-trained knowledge. In this paper, we propose Mutual Information-guided Sparse Tuning (MIST), a plug-and-play method that selectively updates a small subset of PTM parameters, less than 5%, based on sensitivity to mutual information objectives. MIST enables effective task-specific adaptation while preserving generalization. To further reduce interference, we introduce strong sparsity regularization by randomly dropping gradients during tuning, resulting in fewer than 0.5% of parameters being updated per step. Applied before standard freeze-based methods, MIST consistently boosts performance across diverse continual learning benchmarks. Experiments show that integrating our method into multiple baselines yields significant performance gains. Our code is available at https://github.com/zhwhu/MIST.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。