LLM-Enhanced Topical Trend Detection at Snapchat
This work provides the first published end-to-end production system for topical trend detection on short-video platforms, addressing a practical need for social media platforms to maintain content freshness at scale.
Snapchat developed and deployed a large-scale system for detecting emerging topical trends on short-video platforms, integrating multimodal extraction, burst detection, and LLM-based enrichment, achieving high precision in human evaluation over six months and driving measurable improvements in content freshness.
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.