LLM-Enhanced Topical Trend Detection at Snapchat

TL;DR

This paper introduces a large-scale, production-ready system that leverages multimodal data, time-series analysis, and large language models to detect emerging trends on Snapchat, improving content relevance and user engagement.

Contribution

The work presents the first end-to-end system for topical trend detection on short-video platforms, integrating multiple techniques for accurate, scalable trend identification.

Findings

High precision in identifying meaningful trends over six months of evaluation.

System deployment improved content freshness and user experience at global scale.

Effective integration of multimodal data and LLMs for trend detection.

Abstract

Automatic detection of topical trends at scale is both challenging and essential for maintaining a dynamic content ecosystem on social media platforms. In this work, we present a large-scale system for identifying emerging topical trends on Snapchat, one of the world's largest short-video social platforms. Our system integrates multimodal topic extraction, time-series burst detection, and LLM-based consolidation and enrichment to enable accurate and timely trend discovery. To the best of our knowledge, this is the first published end-to-end system for topical trend detection on short-video platforms at production scale. Continuous offline human evaluation over six months demonstrates high precision in identifying meaningful trends. The system has been deployed in production at global scale and applied to downstream surfaces including content ranking and search, driving measurable improvements in content freshness and user experience.