Multi-modal Adaptive Mixture of Experts for Cold-start Recommendation

TL;DR

This paper introduces MAMEX, a novel multimodal mixture of experts framework that dynamically integrates different data modalities to improve cold-start recommendation accuracy and robustness.

Contribution

It proposes a learnable gating mechanism within a Mixture of Experts model for adaptive multimodal data integration in cold-start recommendation tasks.

Findings

MAMEX outperforms existing methods in cold-start scenarios.

The model effectively emphasizes the most informative modalities.

MAMEX maintains robustness when modalities are missing or less relevant.

Abstract

Recommendation systems have faced significant challenges in cold-start scenarios, where new items with a limited history of interaction need to be effectively recommended to users. Though multimodal data (e.g., images, text, audio, etc.) offer rich information to address this issue, existing approaches often employ simplistic integration methods such as concatenation, average pooling, or fixed weighting schemes, which fail to capture the complex relationships between modalities. Our study proposes a novel Mixture of Experts (MoE) framework for multimodal cold-start recommendation, named MAMEX, which dynamically leverages latent representation from different modalities. MAMEX utilizes modality-specific expert networks and introduces a learnable gating mechanism that adaptively weights the contribution of each modality based on its content characteristics. This approach enables MAMEX to emphasize the most informative modalities for each item while maintaining robustness when certain modalities are less relevant or missing. Extensive experiments on benchmark datasets show that MAMEX outperforms state-of-the-art methods in cold-start scenarios, with superior accuracy and adaptability. For reproducibility, the code has been made available on Github https://github.com/L2R-UET/MAMEX.