OmniTrend: Content-Context Modeling for Scalable Social Popularity Prediction

2026-04-28

AI Summary

This work addresses the challenge of social media popularity prediction, which is often hindered by the entanglement of intrinsic content appeal and contextual platform exposure, leading to poor model interpretability and limited cross-platform generalization. To resolve this, the authors propose OmniTrend, a novel framework that explicitly disentangles popularity into the joint effect of content attractiveness and contextual exposure, modeling each component separately before fusing their predictions. The content module leverages multimodal encoding of text, visual, and audio signals, while the context module integrates external cues such as posting time, creator activity, trending topics, and neighborhood statistics. Experimental results demonstrate that OmniTrend achieves robust performance across both image- and video-based platforms, significantly enhancing cross-platform transferability and model interpretability.

Abstract

Predicting social media popularity requires understanding both the intrinsic appeal of content and the external context that determines how it is exposed to users. Existing methods focus on content signals but do not separate them from exposure-related patterns, which causes the learned representations to absorb platform-specific visibility effects and weakens both interpretability and cross-platform transfer. This paper introduces OmniTrend, a unified framework that models popularity as the joint outcome of content attractiveness and contextual exposure. The content module learns cross-modal representations from visual, audio, and textual cues to quantify intrinsic appeal, while the context module estimates exposure from exogenous signals such as posting time, author activity, topical trends, and retrieval-based neighborhood statistics. OmniTrend learns separate predictors for content attractiveness and contextual exposure and integrates them in the final popularity estimate, which makes the role of each factor explicit and supports robust transfer across image and video platforms.

Problem

• social popularity prediction
• content-context separation
• cross-platform transfer
• exposure bias
• intrinsic appeal

Innovation

• content-context modeling
• social popularity prediction
• cross-modal representation
• exposure estimation
• cross-platform transfer