One Size Does Not Fit All: Adaptive Interest Representation Learning for Multi-interest Recommendation
Open Access
- 28 October 2026
- journal article
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Recommender Systems
- Vol. 5 (2) , 1-31
Abstract
Sequential recommendation (SR) aims to predict users’ next interactions by modeling contextual patterns in their historical behavior sequences. Recent advancements in SR introduce multi-interest models that use multiple vectors to represent users’ diverse interests. However, existing methods predefine the number of interest vectors as a hyperparameter, limiting their adaptability to real-world scenarios where the number of interests varies across users. Our experiments show that this fixed-number interest representation suffers from a critical limitation: overfitting for users with simple interests, underfitting for users with complex interests, and a lack of interest prioritization. To address these limitations, we propose A daptive i nterest r epresentation learning for multi-interest Rec ommendation (AirRec), a two-stage framework that adaptively adjusts the number of interest representations for each user’s behavior sequence. Specifically, the first stage focuses on interest foundation construction: to compensate for the lack of auxiliary information, we decompose item co-occurrence data to extract atomic interests, which act as the basic units for interest representation modeling. In the second stage, we conduct user-adaptive interest modeling, retaining only atomic interests relevant to the user’s interaction history to filter out noise and infer the actual number of underlying interests. The selected atomic interests are then aggregated into personalized, adaptively sized interest representations for each user sequence. The entire process is designed for parallel computation, guaranteeing efficiency in both training and inference. Experiments on three benchmark datasets demonstrate that AirRec improves NDCG@20 by an average of 14.94% and NDCG@50 by 24.73% compared to state-of-the-art methods. The source code is available at https://github.com/yaokunliu/AirRec.git.Keywords
Funding Information
- National Science Foundation (CNS-2427070, IIS-2331069, IIS-2202481, IIS-2130263, and CNS-2131622)
This publication has 34 references indexed in Scilit:
- Fundamentals of Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) networkPhysica D: Nonlinear Phenomena, 2020
- Multi-Interest Network with Dynamic Routing for Recommendation at Tmall2019
- Self-Attentive Sequential Recommendation2018
- An Attribute-aware Neural Attentive Model for Next Basket Recommendation2018
- Improving Sequential Recommendation with Knowledge-Enhanced Memory Networks2018
- Sequential User-based Recurrent Neural Network Recommendations2017
- Fusing Similarity Models with Markov Chains for Sparse Sequential Recommendation2016
- Deep Neural Networks for YouTube Recommendations2016
- Ups and Downs2016
- Factorizing personalized Markov chains for next-basket recommendation2010