CLEAR: Null-Space Projection for Cross-Modal De-Redundancy in Multimodal Recommendation

TL;DR

CLEAR is a novel method that reduces cross-modal redundancy in multimodal recommendation systems by projecting features onto a null space, improving recommendation accuracy without altering existing models.

Contribution

The paper introduces a lightweight, plug-and-play approach called CLEAR that explicitly characterizes and suppresses redundant shared components across modalities in recommendation systems.

Findings

Consistently improves recommendation performance across benchmarks.

Effectively reduces cross-modal redundancy without modifying existing architectures.

Enhances utilization of complementary multimodal information.

Abstract

Multimodal recommendation has emerged as an effective paradigm for enhancing collaborative filtering by incorporating heterogeneous content modalities. Existing multimodal recommenders predominantly focus on reinforcing cross-modal consistency to facilitate multimodal fusion. However, we observe that multimodal representations often exhibit substantial cross-modal redundancy, where dominant shared components overlap across modalities. Such redundancy can limit the effective utilization of complementary information, explaining why incorporating additional modalities does not always yield performance improvements. In this work, we propose CLEAR, a lightweight and plug-and-play cross-modal de-redundancy approach for multimodal recommendation. Rather than enforcing stronger cross-modal alignment, CLEAR explicitly characterizes the redundant shared subspace across modalities by modeling cross-modal covariance between visual and textual representations. By identifying dominant shared directions via singular value decomposition and projecting multimodal features onto the complementary null space, CLEAR reshapes the multimodal representation space by suppressing redundant cross-modal components while preserving modality-specific information. This subspace-level projection implicitly regulates representation learning dynamics, preventing the model from repeatedly amplifying redundant shared semantics during training. Notably, CLEAR can be seamlessly integrated into existing multimodal recommenders without modifying their architectures or training objectives. Extensive experiments on three public benchmark datasets demonstrate that explicitly reducing cross-modal redundancy consistently improves recommendation performance across a wide range of multimodal recommendation models.