From Clicks to Intent: Cross-Platform Session Embeddings with LLM-Distilled Taxonomy for Financial Services Recommendations
For financial services recommender systems, this work enables cross-platform intent transfer from anonymous web sessions to authenticated mobile apps, improving personalization while maintaining interpretability.
The paper addresses the gap in using pre-login web clickstreams for post-authentication personalization in financial services. It proposes a framework that combines self-supervised Transformer session embeddings with LLM-distilled intent labels, achieving 1.88% improvement in macro Recall@1 and 13.38% reduction in Log Loss on mobile homepage tile ranking, and outperforming LLM labels by 4.3% micro F1 on conversion prediction.
Sequential user behavior modeling is widely adopted in industrial recommender systems; however, significant gaps remain in financial services, where pre-login web interactions and authenticated in-app experiences differ drastically. Specifically, pre-login web users typically explore new products, whereas logged-in app users focus on account servicing. Due to the challenge of cross-channel entity resolution (e.g., matching anonymous web sessions to authenticated mobile accounts), web-based intent signals remain underutilized for post-authentication personalization. Existing methods for capturing web-based intent are often ad-hoc and narrow, lacking the flexibility to support both quantitative downstream recommendations and qualitative understanding at scale. In this work, we propose a scalable and dual-purpose intent prediction framework for web-based interactions and demonstrate its applicability for personalization. Our approach transforms raw web clickstreams into two outputs: a self-supervised Transformer encodes multi-modal clickstreams into a compact session embedding, while an LLM-based taxonomy generation and distillation pipeline produces interpretable intent labels. Our system demonstrates that self-supervised clickstream representations combined with LLM-distilled taxonomies can jointly serve quantitative tasks and qualitative understanding in production: on the mobile homepage tile ranking task, the session embedding improves macro Recall@1 by 1.88% and reduces Log Loss by 13.38% over production baselines. On the user conversion prediction task, the embedding outperforms the LLM labels by 4.3% on micro F1, while the distillation layer delivers interpretable labels at ultra-low latency with only a 7% performance drop.