AdaM-Rec: Adaptive Modality Routing for Multimodal Recommendation

2026-09-29

AI Summary

This study addresses the imbalance in the importance of visual and textual signals across different scenarios caused by static fusion strategies in multimodal recommendation. To overcome this limitation, this work proposes an agent-driven dynamic modality routing framework based on large language models. Departing from fixed fusion paradigms, the framework dynamically calibrates modality dependencies through a proxy recall task, enabling fine-grained, on-demand invocation of visual and semantic signals. Furthermore, it optimizes personalized queries by integrating structured natural language representations with collaborative filtering enhancement. Extensive experiments on multiple benchmark datasets demonstrate that the proposed approach significantly outperforms existing state-of-the-art methods, validating the effectiveness of adaptively controlling modality dependencies for improving the quality of multimodal recommendation.

Abstract

While recent multimodal recommender systems have demonstrated the effectiveness of incorporating visual and textual information to improve downstream performance, most existing methods rely on static modality fusion, assuming that the relative importance of textual and visual signals remains stable across recommendation scenarios. This design may not fully account for an important variation across recommendation requests: some queries require fine-grained visual cues, whereas others are better served by textual or functional semantics, in which case indiscriminate modality fusion brings in uninformative cues and impairs recommendation quality. To address this, we propose AdaM-Rec, an LLM-based framework for adaptive modality routing in multimodal recommendation, which enables dynamic calibration of reliance on textual and multimodal evidence for user-specific queries. Built on structured natural-language representations of items and user preferences, it estimates modality reliability using proxy recall tasks. Specifically, it generates pseudo-queries that match the granularity of the actual query while pointing to the user's positively interacted items as verifiable proxy targets, evaluating which modality yields better recall performance in analogous scenarios and optimizing the routing strategy in an agentic manner. It then performs routed recall with optimized strategy, enriches results with collaborative items, and ranks candidates by their relevance to both the query and user preferences. Experiments demonstrate that AdaM-Rec delivers strong performance against state-of-the-art baselines, highlighting the effectiveness and broader potential of adaptive control over modality reliance in multimodal recommendation.

Problem

• Multimodal Recommendation
• Adaptive Modality Routing
• Static Modality Fusion
• Dynamic Modality Reliance

Innovation

• Adaptive Modality Routing
• Multimodal Recommendation
• LLM-based Framework
• Proxy Recall Tasks
• Agentic Optimization