GTS-LUM: Reshaping User Behavior Modeling with LLMs in Telecommunications Industry

2025-04-09

AI Summary

In telecom scenarios, user behavior exhibits complex characteristics—including long-term periodicity, multi-granularity temporal patterns, multimodality (structured tabular data + behavioral co-occurrence graphs), and heterogeneous labels—posing significant challenges for unified modeling. To address this, we propose an end-to-end multimodal user behavior modeling framework tailored to telecom applications. Our method introduces: (i) the first LLM-based decoder architecture specifically designed for telecom behavior modeling; (ii) timestamp encoding with explicit temporal awareness; (iii) Q-former–driven cross-modal alignment between tabular and graph modalities; (iv) target-guided historical behavior focusing; and (v) fine-grained temporal embedding. Evaluated on a large-scale industrial telecom dataset, our framework significantly outperforms state-of-the-art baselines (e.g., LLM4Rec) on both behavioral prediction and marketing intervention effect estimation. It achieves strong generalization while maintaining operational practicality, directly supporting critical business decisions such as tariff plan design and precision marketing.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

As telecommunication service providers shifting their focus to analyzing user behavior for package design and marketing interventions, a critical challenge lies in developing a unified, end-to-end framework capable of modeling long-term and periodic user behavior sequences with diverse time granularities, multi-modal data inputs, and heterogeneous labels. This paper introduces GTS-LUM, a novel user behavior model that redefines modeling paradigms in telecommunication settings. GTS-LUM adopts a (multi-modal) encoder-adapter-LLM decoder architecture, enhanced with several telecom-specific innovations. Specifically, the model incorporates an advanced timestamp processing method to handle varying time granularities. It also supports multi-modal data inputs -- including structured tables and behavior co-occurrence graphs -- and aligns these with semantic information extracted by a tokenizer using a Q-former structure. Additionally, GTS-LUM integrates a front-placed target-aware mechanism to highlight historical behaviors most relevant to the target. Extensive experiments on industrial dataset validate the effectiveness of this end-to-end framework and also demonstrate that GTS-LUM outperforms LLM4Rec approaches which are popular in recommendation systems, offering an effective and generalizing solution for user behavior modeling in telecommunications.

Problem

• Modeling long-term and periodic user behavior sequences with diverse time granularities.
• Handling multi-modal data inputs and heterogeneous labels in telecommunication settings.
• Developing a unified, end-to-end framework for user behavior analysis and marketing.

Innovation

• Encoder-adapter-LLM decoder for telecom behavior modeling
• Advanced timestamp processing for varying granularities
• Multi-modal data alignment with Q-former structure