用降维提示调优，让大模型持续学习时省资源还记牢旧知识。
SPARC: Subspace-Aware Prompt Adaptation for Robust Continual Learning in LLMs
- 通过PCA找数据关键特征子空间，只在低维空间调提示。
- 仅更新0.04%参数，任务和领域增量学习都保持高知识保留率。
- 结合LoRA灵活控制算力与精度，适合资源受限场景使用。
我们提出SPARC，一种面向大语言模型的轻量级持续学习框架，通过在低维子空间中进行提示调优，实现高效任务适应。利用主成分分析（PCA）识别训练数据的紧凑子空间，在该子空间优化提示可提升训练效率，聚焦于最相关特征并降低计算开销。由于模型内部结构不变，预训练获得的广泛知识得以完整保留，避免了适应过程中的知识遗忘。该方法在任务增量和领域增量设置下均实现高知识保留率，且仅需微调0.04%的模型参数。进一步结合LoRA技术，增强了对计算资源约束的适应性，可在准确率与训练成本间灵活权衡。SuperGLUE基准测试表明，基于PCA的提示调优配合LoRA，仅使用1%参数即可维持全知识保留并提升性能，验证了该方法在大模型持续学习中的可扩展性与资源高效性。
原文摘要 · Abstract (English)
We propose SPARC, a lightweight continual learning framework for large language models (LLMs) that enables efficient task adaptation through prompt tuning in a lower-dimensional space. By leveraging principal component analysis (PCA), we identify a compact subspace of the training data. Optimizing prompts in this lower-dimensional space enhances training efficiency, as it focuses updates on the most relevant features while reducing computational overhead. Furthermore, since the model's internal structure remains unaltered, the extensive knowledge gained from pretraining is fully preserved, ensuring that previously learned information is not compromised during adaptation. Our method achieves high knowledge retention in both task-incremental and domain-incremental continual learning setups while fine-tuning only 0.04% of the model's parameters. Additionally, by integrating LoRA, we enhance adaptability to computational constraints, allowing for a tradeoff between accuracy and training cost. Experiments on the SuperGLUE benchmark demonstrate that our PCA-based prompt tuning combined with LoRA maintains full knowledge retention while improving accuracy, utilizing only 1% of the model's parameters. These results establish our approach as a scalable and resource-efficient solution for continual learning in LLMs.
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。