Unified Representation Learning for Multi-Intent Diversity and Behavioral Uncertainty in Recommender Systems

2025-09-04

AI Summary

To address the challenges of modeling user intent diversity and behavioral uncertainty in recommender systems, this paper proposes a joint multi-intent–uncertainty representation framework. Methodologically, it employs multiple latent-variable intent vectors to capture fine-grained user interests and introduces an attention-weighted fusion mechanism to integrate long-term preferences with short-term behavioral signals. Behavioral representations are modeled via Gaussian distributions—parameterizing both mean and covariance—to enable Bayesian uncertainty quantification. A unified, end-to-end differentiable objective jointly optimizes intent disentanglement and uncertainty estimation. Extensive experiments on multiple public benchmarks demonstrate significant improvements over state-of-the-art methods in Recall and NDCG. Notably, the framework exhibits markedly enhanced robustness and generalization under cold-start and behavior-perturbation scenarios, validating the effectiveness and practicality of joint intent–uncertainty modeling.

Technology Category

• Reasoning under Uncertainty: Uncertainty Representations
• Machine Learning: Calibration & Uncertainty Quantification
• Data Mining & Knowledge Management: Recommender Systems

Application Category

• User Modeling, Personalization and Recommendation: Psychology-informed user models and recommender systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

This paper addresses the challenge of jointly modeling user intent diversity and behavioral uncertainty in recommender systems. A unified representation learning framework is proposed. The framework builds a multi-intent representation module and an uncertainty modeling mechanism. It extracts multi-granularity interest structures from user behavior sequences. Behavioral ambiguity and preference fluctuation are captured using Bayesian distribution modeling. In the multi-intent modeling part, the model introduces multiple latent intent vectors. These vectors are weighted and fused using an attention mechanism to generate semantically rich representations of long-term user preferences. In the uncertainty modeling part, the model learns the mean and covariance of behavior representations through Gaussian distributions. This reflects the user's confidence in different behavioral contexts. Next, a learnable fusion strategy is used to combine long-term intent and short-term behavior signals. This produces the final user representation, improving both recommendation accuracy and robustness. The method is evaluated on standard public datasets. Experimental results show that it outperforms existing representative models across multiple metrics. It also demonstrates greater stability and adaptability under cold-start and behavioral disturbance scenarios. The approach alleviates modeling bottlenecks faced by traditional methods when dealing with complex user behavior. These findings confirm the effectiveness and practical value of the unified modeling strategy in real-world recommendation tasks.

Problem

• Modeling user intent diversity and behavioral uncertainty
• Extracting multi-granularity interest structures from sequences
• Improving recommendation accuracy and robustness through unified representation

Innovation

• Multi-intent representation module with attention fusion
• Bayesian uncertainty modeling using Gaussian distributions
• Learnable fusion of long-term and short-term signals