STAIR: Manipulating Collaborative and Multimodal Information for E-Commerce Recommendation

TL;DR

This paper introduces STAIR, a novel graph convolution method that effectively combines collaborative and multimodal information for e-commerce recommendations, addressing modality erasure and forgetting issues to improve performance.

Contribution

The paper proposes STAIR, a new stepwise graph convolution approach that preserves multimodal features and mitigates forgetting, achieving state-of-the-art results in e-commerce recommendation tasks.

Findings

Achieves state-of-the-art performance on three datasets.

Effectively preserves multimodal information during training.

Requires minimal computational and memory resources.

Abstract

While the mining of modalities is the focus of most multimodal recommendation methods, we believe that how to fully utilize both collaborative and multimodal information is pivotal in e-commerce scenarios where, as clarified in this work, the user behaviors are rarely determined entirely by multimodal features. In order to combine the two distinct types of information, some additional challenges are encountered: 1) Modality erasure: Vanilla graph convolution, which proves rather useful in collaborative filtering, however erases multimodal information; 2) Modality forgetting: Multimodal information tends to be gradually forgotten as the recommendation loss essentially facilitates the learning of collaborative information. To this end, we propose a novel approach named STAIR, which employs a novel STepwise grAph convolution to enable a co-existence of collaborative and multimodal Information in e-commerce Recommendation. Besides, it starts with the raw multimodal features as an initialization, and the forgetting problem can be significantly alleviated through constrained embedding updates. As a result, STAIR achieves state-of-the-art recommendation performance on three public e-commerce datasets with minimal computational and memory costs. Our code is available at https://github.com/yhhe2004/STAIR.