From Events to Trending: A Multi-Stage Hotspots Detection Method Based on Generative Query Indexing

TL;DR

This paper introduces a multi-stage hotspots detection framework for trending queries in conversational systems, combining offline event-based indexing with online retrieval to improve detection accuracy and user satisfaction.

Contribution

It presents a novel multi-stage detection framework that integrates static event indexing with real-time retrieval, tailored for dialogue systems.

Findings

Significantly outperforms baseline methods in offline evaluations.

Achieves 27% improvement in user satisfaction in online tests.

Effective cold-start strategy with single-recall module.

Abstract

LLM-based conversational systems have become a popular gateway for information access, yet most existing chatbots struggle to handle news-related trending queries effectively. To improve user experience, an effective trending query detection method is urgently needed to enable differentiated processing of such target traffic. However, current research on trending detection tailored to the dialogue system scenario remains largely unexplored, and methods designed for traditional search engines often underperform in conversational contexts due to radically distinct query distributions and expression patterns. To fill this gap, we propose a multi-stage framework for trending detection, which achieves systematic optimization from both offline generation and online identification perspectives. Specifically, our framework first exploits selected hot events to generate index queries, establishing a key bridge between static events and dynamic user queries. It then employs a retrieval matching mechanism for real-time online detection of trending queries, where we introduce a cascaded recall and ranking architecture to balance detection efficiency and accuracy. Furthermore, to better adapt to the practical application scenario, our framework adopts a single-recall module as a cold-start strategy to collect online data for fine-tuning the reranker. Extensive experiments demonstrate that our framework significantly outperforms baseline methods in both offline evaluations and online A/B tests, and user satisfaction is relatively improved by 27\% in terms of positive-negative feedback ratio.