HyMoERec: Hybrid Mixture-of-Experts for Sequential Recommendation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Kunrong, Sun, Zhu, Lim, Kwan Hui
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915607188865024
author Li, Kunrong
Sun, Zhu
Lim, Kwan Hui
author_facet Li, Kunrong
Sun, Zhu
Lim, Kwan Hui
contents We propose HyMoERec, a novel sequential recommendation framework that addresses the limitations of uniform Position-wise Feed-Forward Networks in existing models. Current approaches treat all user interactions and items equally, overlooking the heterogeneity in user behavior patterns and diversity in item complexity. HyMoERec initially introduces a hybrid mixture-of-experts architecture that combines shared and specialized expert branches with an adaptive expert fusion mechanism for the sequential recommendation task. This design captures diverse reasoning for varied users and items while ensuring stable training. Experiments on MovieLens-1M and Beauty datasets demonstrate that HyMoERec consistently outperforms state-of-the-art baselines.
format Preprint
id arxiv_https___arxiv_org_abs_2511_06388
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle HyMoERec: Hybrid Mixture-of-Experts for Sequential Recommendation
Li, Kunrong
Sun, Zhu
Lim, Kwan Hui
Information Retrieval
Artificial Intelligence
We propose HyMoERec, a novel sequential recommendation framework that addresses the limitations of uniform Position-wise Feed-Forward Networks in existing models. Current approaches treat all user interactions and items equally, overlooking the heterogeneity in user behavior patterns and diversity in item complexity. HyMoERec initially introduces a hybrid mixture-of-experts architecture that combines shared and specialized expert branches with an adaptive expert fusion mechanism for the sequential recommendation task. This design captures diverse reasoning for varied users and items while ensuring stable training. Experiments on MovieLens-1M and Beauty datasets demonstrate that HyMoERec consistently outperforms state-of-the-art baselines.
title HyMoERec: Hybrid Mixture-of-Experts for Sequential Recommendation
topic Information Retrieval
Artificial Intelligence
url https://arxiv.org/abs/2511.06388