TimeRoute: Time-Aware Modality Routing and Diffusion for Multi-Modal Recommendation
This work provides an incremental improvement for multi-modal recommender systems by dynamically adjusting modality contributions based on temporal context, which is relevant for e-commerce and content platforms.
The paper addresses the challenge of time-varying usefulness of different modalities in multi-modal recommenders. It proposes TimeRoute, a diffusion-based recommender that uses a temporal-aware modal router to generate personalized modality distributions and a dual-stream denoising head to suppress outdated modality signals. Experiments show improvements of up to 9.8% in Recall@K, Precision@K, and NDCG@K on TikTok, Amazon-Baby, and Amazon-Sports datasets.
Multi-modal recommenders fuse collaborative signals with item modalities such as text, images, and audio, but the usefulness of each drifts over time and at different rates. For example, chocolate purchases typically guided by textual ingredient cues can shift toward visual packaging and ambient audio around Valentine's Day. This modality time-scale mismatch gives rise to two coupled challenges: (1) users require different modality proportions across temporal contexts, and (2) less relevant modalities are more likely to introduce outdated or misleading signals into the recommender. We address both challenges within a unified diffusion-based recommender, TimeRoute. A temporal-aware modal router maps each user's aggregated behavioral features to a personalized modality distribution, replacing the globally shared fusion weights used in prior work. The diffusion-based graph reconstructor is then conditioned on the same temporal profile through Feature-wise Linear Modulation (FiLM) with dual-stream long- and short-term denoising heads, suppressing outdated modality edges before they enter the propagation graph. Experiments on TikTok, Amazon-Baby, and Amazon-Sports demonstrate consistent improvements of up to 9.8\% in Recall@K, Precision@K, and NDCG@K over strong baselines across 10-seed paired tests. Code is available at https://anonymous.4open.science/r/TimeRoute.