AdaM-Rec：面向多模态推荐的自适应模态路由

AI 摘要

研究者提出 AdaM-Rec，一个基于 LLM 的多模态推荐框架，用于自适应地路由文本与视觉模态的依赖程度。该方法针对现有系统静态融合模态、忽略查询级差异的问题，利用用户历史正向交互物品构建代理召回任务，生成伪查询来评估哪种模态更能捕捉用户意图，并以 agentic 方式优化路由策略。实验表明其性能优于当前最优基线，代码已开源。

正文节选

AdaM-Rec: Adaptive Modality Routing for 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:

发布时间：2026-10-01 12:00

抓取时间：2026-10-01 12:55

来源机构：arXiv