Modeling User Behavior from Adaptive Surveys with Supplemental Context

TL;DR

This paper introduces LANTERN, a modular neural architecture that enhances user behavior modeling by integrating adaptive survey responses with external contextual signals, improving prediction accuracy and scalability.

Contribution

The work presents a novel modular architecture that effectively fuses survey data with supplemental context, maintaining survey primacy and enabling scalable, multi-modal behavior modeling.

Findings

LANTERN outperforms survey-only models in multi-label response prediction.

Selective gating and late fusion improve model robustness and relevance.

Modularity allows easy integration of new data encoders.

Abstract

Modeling user behavior is critical across many industries where understanding preferences, intent, or decisions informs personalization, targeting, and strategic outcomes. Surveys have long served as a classical mechanism for collecting such behavioral data due to their interpretability, structure, and ease of deployment. However, surveys alone are inherently limited by user fatigue, incomplete responses, and practical constraints on their length making them insufficient for capturing user behavior. In this work, we present LANTERN (Late-Attentive Network for Enriched Response Modeling), a modular architecture for modeling user behavior by fusing adaptive survey responses with supplemental contextual signals. We demonstrate the architectural value of maintaining survey primacy through selective gating, residual connections and late fusion via cross-attention, treating survey data as the primary signal while incorporating external modalities only when relevant. LANTERN outperforms strong survey-only baselines in multi-label prediction of survey responses. We further investigate threshold sensitivity and the benefits of selective modality reliance through ablation and rare/frequent attribute analysis. LANTERN's modularity supports scalable integration of new encoders and evolving datasets. This work provides a practical and extensible blueprint for behavior modeling in survey-centric applications.