提出自适应回放机制，缓解大模型持续学习中的遗忘问题。
MSSR: Memory-Aware Adaptive Replay for Continual LLM Fine-Tuning
- 根据记忆强度动态选择回放样本，避免固定策略浪费资源。
- 在11个任务上显著优于现有方法，尤其在推理类任务提升明显。
- 适合需要长期迭代更新的智能系统，如客服、知识库维护。
大型语言模型在动态环境中持续微调的重要性日益凸显。尽管其具备快速获取新知识的能力，但也会面临灾难性遗忘问题。现有基于回放的方法或依赖启发式规则、缓解有限，或虽有效却带来高昂计算开销。本文受连续微调中保留规律启发，提出内存感知的采样与调度回放框架（MSSR），通过估计样本级记忆强度，并自适应安排回放周期，以减轻遗忘并保持快速适应。在三个骨干模型和11个顺序任务上的实验表明，MSSR始终优于当前最优回放基线，在推理密集型和多项选择类基准上表现尤为突出。
原文摘要 · Abstract (English)
Continual fine-tuning of large language models (LLMs) is becoming increasingly crucial as these models are deployed in dynamic environments where tasks and data distributions evolve over time. While strong adaptability enables rapid acquisition of new knowledge, it also exposes LLMs to catastrophic forgetting, where previously learned skills degrade during sequential training. Existing replay-based strategies, such as fixed interleaved replay, accuracy-supervised, and loss-driven scheduling, remain limited: some depend on heuristic rules and provide only partial mitigation of forgetting, while others improve performance but incur substantial computational overhead. Motivated by retention dynamics under sequential fine-tuning, we propose Memory-Inspired Sampler and Scheduler Replay (MSSR), an experience replay framework that estimates sample-level memory strength and schedules rehearsal at adaptive intervals to mitigate catastrophic forgetting while maintaining fast adaptation. Extensive experiments across three backbone models and 11 sequential tasks show that MSSR consistently outperforms state-of-the-art replay baselines, with particularly strong gains on reasoning-intensive and multiple-choice benchmarks.
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。