Tricolore: Multi-Behavior User Profiling for Enhanced Candidate Generation in Recommender Systems

TL;DR

Tricolore introduces a multi-vector learning framework for recommender systems that captures diverse user behaviors, improving candidate generation, diversity, and cold-start performance across various platforms.

Contribution

It presents a novel multi-vector approach with adaptive multi-task learning and behavior-wise fusion, addressing limitations of single-vector models in multi-behavior user profiling.

Findings

Enhanced recommendation accuracy across multiple datasets

Improved cold-start user performance

Balanced diversity and popularity in recommendations

Abstract

Online platforms aggregate extensive user feedback across diverse behaviors, providing a rich source for enhancing user engagement. Traditional recommender systems, however, typically optimize for a single target behavior and represent user preferences with a single vector, limiting their ability to handle multiple important behaviors or optimization objectives. This conventional approach also struggles to capture the full spectrum of user interests, resulting in a narrow item pool during candidate generation. To address these limitations, we present Tricolore, a versatile multi-vector learning framework that uncovers connections between different behavior types for more robust candidate generation. Tricolore's adaptive multi-task structure is also customizable to specific platform needs. To manage the variability in sparsity across behavior types, we incorporate a behavior-wise multi-view fusion module that dynamically enhances learning. Moreover, a popularity-balanced strategy ensures the recommendation list balances accuracy with item popularity, fostering diversity and improving overall performance. Extensive experiments on public datasets demonstrate Tricolore's effectiveness across various recommendation scenarios, from short video platforms to e-commerce. By leveraging a shared base embedding strategy, Tricolore also significantly improves the performance for cold-start users. The source code is publicly available at: https://github.com/abnering/Tricolore.