Saved in:
Bibliographic Details
Main Authors: Wang, Junting, Guo, Chenghuan, Yang, Jiao, Guo, Yanhui, Sundaram, Hari, Gao, Yan
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2507.22268
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915974601506816
author Wang, Junting
Guo, Chenghuan
Yang, Jiao
Guo, Yanhui
Sundaram, Hari
Gao, Yan
author_facet Wang, Junting
Guo, Chenghuan
Yang, Jiao
Guo, Yanhui
Sundaram, Hari
Gao, Yan
contents We study the problem of inferring substitutable and complementary items, which underpins applications such as alternative and follow-up purchase suggestions. Existing approaches typically learn from behavior-derived item-item associations using GNNs or leverage item content alone. However, these methods often overlook two key challenges: (i) user behaviors (e.g., co-view/co-purchase) only provide noisy weak supervision, and (ii) behavior signals are long-tailed, leaving many items with sparse associations. We propose MMSC, a self-supervised multi-modal relational representation learning framework that combines a multi-modal foundation model adapted to encode item metadata and a self-supervised denoising module that learns relationship-aware representations from noisy user behaviors, unified by a hierarchical aggregation mechanism. We further use LLM-assisted supervision to mitigate noise in behavior-derived supervision during training. Experiments on five real-world datasets show that MMSC consistently outperforms existing baselines by 26.1% for substitutable and 39.2% for complementary item inference, while remaining effective for cold-start items. We share our code for reproducibility.
format Preprint
id arxiv_https___arxiv_org_abs_2507_22268
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Multi-modal Relational Item Representation Learning for Inferring Substitutable and Complementary Items
Wang, Junting
Guo, Chenghuan
Yang, Jiao
Guo, Yanhui
Sundaram, Hari
Gao, Yan
Information Retrieval
Artificial Intelligence
We study the problem of inferring substitutable and complementary items, which underpins applications such as alternative and follow-up purchase suggestions. Existing approaches typically learn from behavior-derived item-item associations using GNNs or leverage item content alone. However, these methods often overlook two key challenges: (i) user behaviors (e.g., co-view/co-purchase) only provide noisy weak supervision, and (ii) behavior signals are long-tailed, leaving many items with sparse associations. We propose MMSC, a self-supervised multi-modal relational representation learning framework that combines a multi-modal foundation model adapted to encode item metadata and a self-supervised denoising module that learns relationship-aware representations from noisy user behaviors, unified by a hierarchical aggregation mechanism. We further use LLM-assisted supervision to mitigate noise in behavior-derived supervision during training. Experiments on five real-world datasets show that MMSC consistently outperforms existing baselines by 26.1% for substitutable and 39.2% for complementary item inference, while remaining effective for cold-start items. We share our code for reproducibility.
title Multi-modal Relational Item Representation Learning for Inferring Substitutable and Complementary Items
topic Information Retrieval
Artificial Intelligence
url https://arxiv.org/abs/2507.22268