CRAFT：面向持续学习的遗忘感知干预式适应
CRAFT: Forgetting-Aware Intervention-Based Adaptation for Continual Learning
AI总结：
CRAFT通过在隐藏表示上学习低秩干预，避免更新模型权重，通过输出分布分歧路由任务、KL散度正则化和合并干预，减少遗忘并提升性能。
摘要：
大型语言模型（LLMs）可通过微调获得新能力，但持续适应常导致灾难性遗忘。我们提出CRAFT，一种持续学习框架，通过学习隐藏表示上的低秩干预避免更新模型权重。CRAFT分为三个阶段：首先根据输出分布分歧将每个任务路由到相似任务组；然后使用KL散度对齐组的先验状态，直接控制遗忘并决定收敛；最后用相同KL信号将更新任务的干预合并到共享表示中。此设计通过单一KL目标统一了路由、正则化和合并。CRAFT在多个基准和模型规模上优于强LoRA方法，同时对任务顺序鲁棒。这些结果表明，通过输出空间分歧引导的表示空间适应，为LLM的持续学习提供了可扩展且原理化的方案。
英文摘要：
Large language models (LLMs) can acquire new capabilities through fine-tuning, but continual adaptation often leads to catastrophic forgetting. We propose CRAFT, a continual learning framework that avoids updating model weights by instead learning low-rank interventions on hidden representations. CRAFT proceeds in three stages: it first routes each task to a group of similar tasks based on output-distribution divergence; it then fine-tunes the model using a Kullback-Leibler (KL) divergence against the group's prior state, which directly controls forgetting and determines convergence; finally, it merges interventions for the updated task into the shared representation using the same KL signal. This design unifies routing, regularization, and merging through a single KL-based objective. CRAFT improves overall performance and reduces forgetting compared to strong LoRA-based approaches across multiple benchmarks and model scales, while remaining robust to task ordering. These results suggest that controlling adaptation in representation space, guided by output-space divergence, provides a scalable and principled approach to continual learning in LLMs.
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。
