改进Transformer自注意力机制，提升日志监控的准确与效率
DeepFilter: A Transformer-style Framework for Accurate and Efficient Process Monitoring
- 重构自注意力机制，更好理解日志语义上下文
- 在保持精度的同时显著降低计算开销
- 适合需要高效精准监控的日志分析项目
过程监控任务对准确性和效率要求极高。现有基于Transformer的方法依赖自注意力进行时间特征融合，但在准确理解语义上下文和高效处理监控日志方面存在局限，难以满足过程监控需求。为此，我们提出DeepFilter，通过改进自注意力机制，在提升准确性的同时增强处理效率。该方法结构简洁、适用性强，可作为过程监控领域新项目启动或现有系统优化的有力基准。
原文摘要 · Abstract (English)
The process monitoring task is characterized by stringent demands for accuracy and efficiency. Current transformer-based methods, characterized by self-attention for temporal fusion, exhibit limitations in accurately understanding the semantic context and efficiently processing monitoring logs, rendering them inadequate for process monitoring. To address these limitations, we introduce DeepFilter, which revises the self-attention mechanism to improve both accuracy and efficiency. As a straightforward yet versatile approach, DeepFilter provides an instrumental baseline for practitioners in process monitoring, whether initiating new projects or enhancing existing capabilities.
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。