RAILS：面向大规模场景的检索增强增量式LLM聚类

中文摘要 AI 辅助

在生产规模下使用大型语言模型（LLM）作为聚类器面临挑战：提示词无法容纳整个标签空间，且逐文档串行处理无法满足真实工作负载所需的吞吐量。我们提出RAILS，一种检索增强的增量式LLM聚类器，它将聚类转化为对不断增长的标签池的简单循环，并通过文档批处理和有界并发实现扩展。在六个公开基准测试上，RAILS平均超过了先前最强的LLM聚类方法，将准确率从51.2%提升至59.3%，NMI从67.2%提升至74.8%，ARI从45.4%提升至54.7%。我们还报告了来自SaaS工单主题发现管道的生产部署证据，在该管道中，RAILS已取代传统的HDBSCAN阶段，提供了更高的聚类质量、透明的提示驱动控制以及有状态的增量操作。

英文摘要

Using a Large Language Model (LLM) as the clusterer at production scale is hard: prompts cannot hold the entire label space, and per-document serial processing does not deliver the throughput real workloads require. We present RAILS, a retrieval-augmented incremental LLM clusterer that turns clustering into a simple loop over a growing label pool and scales through document batching with bounded concurrency. On six public benchmarks RAILS exceeds the strongest prior LLM-clustering method on average, lifting accuracy from 51.2% to 59.3%, NMI from 67.2% to 74.8%, and ARI from 45.4% to 54.7%. We further report production-deployment evidence from a SaaS ticket-topic-discovery pipeline, where RAILS has replaced a traditional HDBSCAN stage with higher clustering quality, transparent prompt-driven control, and stateful incremental operation.

发表机构

• Zendesk

机构由 AI 辅助整理，请以论文原文为准。