改进稀疏记忆微调
Improving Sparse Memory Finetuning
AI总结：
本文提出稀疏记忆微调方法，通过局部更新显式内存层参数，解决持续学习中的灾难性遗忘问题，并在消费级硬件上验证了其有效性。
摘要：
大型语言模型（LLMs）通常在训练后是静态的，但实际应用需要持续适应新知识而不损害现有能力。标准更新方法如全微调或参数高效方法（如LoRA）面临根本性权衡：灾难性遗忘。稀疏记忆微调（SMF）通过在显式内存层中局部更新少量参数，提供了一种有前途的替代方案。本文提出一个开源流程，将现有预训练模型（Qwen-2.5-0.5B）与稀疏内存模块结合，实现有效的持续学习。我们通过基于Kullback-Leibler（KL）散度的理论指导的槽选择机制扩展了先前工作，该机制优先更新相对于背景分布信息上“令人惊讶”的令牌。实验表明，我们的改进模型能以最小的遗忘率获取新事实知识，验证了稀疏更新假设在实际环境中的有效性。
英文摘要：
Large Language Models (LLMs) are typically static after training, yet real-world applications require continual adaptation to new knowledge without degrading existing capabilities. Standard approaches to updating models, like full finetuning or parameter-efficient methods (e.g., LoRA), face a fundamental trade-off: catastrophic forgetting. They modify shared dense representations, causing interference across tasks. Sparse Memory Finetuning (SMF) offers a promising alternative by localizing updates to a small subset of parameters in explicit memory layers. In this work, we present an open-source pipeline to retrofit existing pretrained models (Qwen-2.5-0.5B) with sparse memory modules, enabling effective continual learning on consumer hardware. We extend prior work by introducing a theoretically grounded slot-selection mechanism based on Kullback-Leibler (KL) divergence, which prioritizes memory updates for informationally "surprising" tokens relative to a background distribution. Our experiments demonstrate that our retrofitted models can acquire new factual knowledge with minimal forgetting of held-out capabilities, validating the sparse update hypothesis in a practical setting.
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。
