让大模型推理时动态学习新知识，提升复杂任务的适应能力。
A Lightweight Framework for Trigger-Guided LoRA-Based Self-Adaptation in LLMs
- 用触发器实时检测推理错误，驱动动态微调。
- 通过流式聚类与相似性合并，稳定识别异常样本。
- 基于LoRA的自适应更新机制，实现知识保留与高效优化。
大型语言模型在推理阶段无法持续适应和学习新数据。为解决这一问题，我们提出将复杂推理任务分解为原子子任务，并引入SAGE框架——一种基于触发器引导的动态微调方法，支持推理时的自适应更新。SAGE包含三个核心组件：(1) 触发器模块，通过多种评估指标实时检测推理失败；(2) 触发缓冲模块，采用流式聚类（HDBSCAN）对异常样本进行聚类，经稳定性检查后进行基于相似性的合并；(3) LoRA存储模块，通过适配器池动态优化参数更新，实现知识保留。实验表明，SAGE在推理过程中通过动态知识更新，在原子推理子任务上展现出优异的准确率、鲁棒性和稳定性。
原文摘要 · Abstract (English)
Large language models are unable to continuously adapt and learn from new data during reasoning at inference time. To address this limitation, we propose that complex reasoning tasks be decomposed into atomic subtasks and introduce SAGE, a trigger-guided dynamic fine-tuning framework that enables adaptive updates during reasoning at inference time. SAGE consists of three key components: (1) a Trigger module that detects reasoning failures through multiple evaluation metrics in real time; (2) a Trigger Buffer module that clusters anomaly samples using a streaming clustering process with HDBSCAN, followed by stability checks and similarity-based merging; and (3) a Lora Store module that dynamically optimizes parameter updates with an adapter pool for knowledge retention. Evaluation results show that SAGE demonstrates excellent accuracy, robustness, and stability on the atomic reasoning subtask through dynamic knowledge updating during test time.
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。