Real-Time Trend Prediction via Continually-Aligned LLM Query Generation

TL;DR

This paper presents RTTP, a real-time trend prediction system using continual learning LLMs to generate search queries from news content, enabling early detection of emerging trends in low-traffic environments.

Contribution

It introduces a novel framework combining LLM-generated queries with a continual learning approach to improve early trend detection and adapt without forgetting.

Findings

+91.4% improvement in tail-trend detection precision@500

+19% query generation accuracy over industry baselines

Stable performance after multi-week online training

Abstract

Trending news detection in low-traffic search environments faces a fundamental cold-start problem, where a lack of query volume prevents systems from identifying emerging or long-tail trends. Existing methods relying on keyword frequency or query spikes are inherently slow and ineffective in these sparse settings, lagging behind real-world shifts in attention. We introduce RTTP, a novel Real-Time Trending Prediction framework that generates search queries directly from news content instead of waiting for users to issue them. RTTP leverages a continual learning LLM (CL-LLM) that converts posts into search-style queries and scores them using engagement strength + creator authority, enabling early trend surfacing before search volume forms. To ensure adaptation without degrading reasoning, we propose Mix-Policy DPO, a new preference-based continual learning approach that combines on-policy stability with off-policy novelty to mitigate catastrophic forgetting during model upgrades. Deployed at production scale on Facebook and Meta AI products, RTTP delivers +91.4% improvement in tail-trend detection precision@500 and +19% query generation accuracy over industry baselines, while sustaining stable performance after multi-week online training. This work demonstrates that LLM-generated synthetic search signals, when aligned and continually updated, unlock timely trend understanding in low-traffic search environments.