AEN通过核密度估计实现无需重训练的实时文本分类，适合边缘设备。
Adaptable Embeddings Network (AEN)
- 用核密度估计构建双编码器，支持运行时动态调整分类标准。
- 在合成数据上表现媲美大10倍的自回归模型，且不依赖重训练。
- 预处理缓存条件嵌入，特别适合边缘计算与实时监控场景。
现代语言模型在文本分类中广泛应用，但计算开销巨大。为低资源环境（尤其是边缘设备）提供高效计算的分类模型势在必行。我们提出适配嵌入网络（Adaptable Embeddings Networks, AEN），一种基于核密度估计（KDE）的新型双编码器架构。该架构可在不重新训练的情况下实现实时分类标准调整，且为非自回归设计。通过大量合成数据实验，我们证明AEN在性能上可媲美甚至优于尺寸大一个数量级的自回归模型。其可预处理并缓存条件嵌入的特性，使其特别适用于边缘计算应用与实时监控系统。
原文摘要 · Abstract (English)
Modern day Language Models see extensive use in text classification, yet this comes at significant computational cost. Compute-effective classification models are needed for low-resource environments, most notably on edge devices. We introduce Adaptable Embeddings Networks (AEN), a novel dual-encoder architecture using Kernel Density Estimation (KDE). This architecture allows for runtime adaptation of classification criteria without retraining and is non-autoregressive. Through thorough synthetic data experimentation, we demonstrate our model outputs comparable and in certain cases superior results to that of autoregressive models an order of magnitude larger than AEN's size. The architecture's ability to preprocess and cache condition embeddings makes it ideal for edge computing applications and real-time monitoring systems.
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。