Modality-Aware Identity Construction and Counterfactual Structure Learning for ID-Free Multimodal Recommendation

TL;DR

This paper introduces MAIL, a novel ID-free multimodal recommendation method that dynamically constructs content-aware identities and mitigates popularity bias through counterfactual structure learning, improving recommendation accuracy.

Contribution

The paper proposes a new modality-aware identity construction module and a counterfactual structure learning paradigm for ID-free multimodal recommendation, addressing static identity representations and popularity bias issues.

Findings

MAIL achieves 7.81% higher Recall@10 on average.

MAIL improves NDCG@10 by 12.81% on average.

Extensive experiments validate the effectiveness of MAIL.

Abstract

Multimodal recommendation has attracted extensive attention by leveraging heterogeneous modality information to alleviate data sparsity and improve recommendation accuracy. Existing methods have attempted to replace ID embeddings with multimodal features and have achieved promising preliminary results. However, these methods still exhibit the following two limitations: (1) the reconstructed ID representations remain relatively static and fail to fully exploit multimodal semantics; and (2) the graph learning process is insufficient in mining latent long-tail semantic relations and is easily affected by popularity bias. To address these issues, we propose a novel method named Modality-Aware Identity Construction and Counterfactual Structure Learning for ID-free Multimodal Recommendation (MAIL). Specifically, we design a modality-aware identity construction module that dynamically modulates positional encodings with multimodal semantics to construct content-aware ID-free identity representations. Then, we propose a counterfactual structure learning paradigm that mines low-exposure semantic neighbors via popularity penalization and alleviates popularity bias. Extensive experiments are conducted on five public Amazon datasets. Experimental results show that MAIL achieves average improvements of 7.81% in Recall@10 and 12.81% in NDCG@10 compared with the baseline models. Our code is available at https://github.com/HubuKG/MAIL.