🤖 AI Summary
This work addresses the challenge of accurately and promptly identifying emerging topic trends on short-video platforms by proposing and deploying the first end-to-end topic trend detection system. The system integrates multimodal topic extraction, time-series burst detection, and the semantic understanding and generation capabilities of large language models (LLMs) to enable efficient and dynamic trend discovery and aggregation. Supported by a large-scale distributed architecture, the system underwent six months of manual evaluation, demonstrating high accuracy, and has since been globally deployed. It significantly enhances content freshness and user experience and has been successfully integrated into critical downstream applications such as content ranking and search.
📝 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.