DATA：基于分解注意力的免复述持续学习任务适配
DATA: Decomposed Attention-based Task Adaptation for Rehearsal-Free Continual Learning
AI总结：
针对持续学习中平衡模型可塑性与灾难性遗忘的挑战，提出DATA方法，通过分解注意力任务适配及动态权重调整策略，在免复述情况下实现新任务学习与旧知识保留的平衡，实验性能达最先进水平。
摘要：
持续学习（CL）对于大型语言模型（LLMs）适应不断变化的现实需求至关重要，但它们容易受到灾难性遗忘（CF）的影响。传统CF解决方案依赖昂贵的数据复述，而近期免复述方法采用基于模型和基于正则化的策略来解决该问题。然而，这些方法常忽略模型的可塑性，而可塑性对新学习任务的最优性能至关重要。因此，CL的关键挑战是平衡可塑性保留与CF缓解。为应对这一挑战，我们提出基于分解注意力的任务适配（DATA），使用高秩和低秩任务适配器（如LoRAs）显式解耦并学习任务特定和任务共享知识。对于新任务，DATA根据适配器与先前任务的相关性和差异性动态调整不同秩适配器的权重，使模型在有效保留已有知识的同时获取新任务特定技能。具体而言，我们实现了分解组件加权策略，包含可学习组件，共同生成基于注意力的权重，让模型整合并利用每个DATA中的多样知识。在三个广泛使用的基准上的大量实验表明，我们的方法达到了最先进的性能。值得注意的是，我们的方法通过扩展可学习组件并在训练迭代中采用随机恢复，显著增强了模型可塑性并缓解了CF。
英文摘要：
Continual learning (CL) is essential for Large Language Models (LLMs) to adapt to evolving real-world demands, yet they are susceptible to catastrophic forgetting (CF). While traditional CF solutions rely on expensive data rehearsal, recent rehearsal-free methods employ model-based and regularization-based strategies to address this issue. However, these approaches often neglect the model's plasticity, which is crucial to achieving optimal performance on newly learned tasks. Consequently, a key challenge in CL is striking a balance between preserving plasticity and mitigating CF. To tackle this challenge, we propose the $\textbf{D}$ecomposed $\textbf{A}$ttention-based $\textbf{T}$ask $\textbf{A}$daptation (DATA), which explicitly decouples and learns both task-specific and task-shared knowledge using high-rank and low-rank task adapters (e.g., LoRAs). For new tasks, DATA dynamically adjusts the weights of adapters of different ranks based on their relevance and distinction from previous tasks, allowing the model to acquire new task-specific skills while effectively retaining previously learned knowledge. Specifically, we implement a decomposed component weighting strategy comprising learnable components that collectively generate attention-based weights, allowing the model to integrate and utilize diverse knowledge from each DATA. Extensive experiments on three widely used benchmarks demonstrate that our proposed method achieves state-of-the-art performance. Notably, our approach significantly enhances model plasticity and mitigates CF by extending learnable components and employing stochastic restoration during training iterations.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
