Title: IRA: Adaptive Interest-aware Representation and Alignment for Personalized Multi-interest Retrieval Show Chinese title

Title: IRA：自适应兴趣感知表示与对齐用于个性化多兴趣检索

Abstract: Online community platforms require dynamic personalized retrieval and recommendation that can continuously adapt to evolving user interests and new documents. However, optimizing models to handle such changes in real-time remains a major challenge in large-scale industrial settings. To address this, we propose the Interest-aware Representation and Alignment (IRA) framework, an efficient and scalable approach that dynamically adapts to new interactions through a cumulative structure. IRA leverages two key mechanisms: (1) Interest Units that capture diverse user interests as contextual texts, while reinforcing or fading over time through cumulative updates, and (2) a retrieval process that measures the relevance between Interest Units and documents based solely on semantic relationships, eliminating dependence on click signals to mitigate temporal biases. By integrating cumulative Interest Unit updates with the retrieval process, IRA continuously adapts to evolving user preferences, ensuring robust and fine-grained personalization without being constrained by past training distributions. We validate the effectiveness of IRA through extensive experiments on real-world datasets, including its deployment in the Home Section of NAVER's CAFE, South Korea's leading community platform. Show Chinese abstract

Abstract: 在线社区平台需要能够持续适应不断变化的用户兴趣和新文档的动态个性化检索和推荐。然而，在大规模工业环境中实时优化模型以应对这些变化仍然是一项重大挑战。 为了解决这个问题，我们提出了兴趣感知表示与对齐（IRA）框架，这是一种高效且可扩展的方法，通过累积结构动态适应新的交互。 IRA 利用两个关键机制：(1) 兴趣单元，它们捕获多样化的用户兴趣作为上下文文本，同时通过累积更新随着时间推移而增强或减弱；(2) 一种检索过程，该过程基于语义关系来衡量兴趣单元与文档之间的相关性，从而消除对点击信号的依赖以减轻时间偏差。 通过将累积的兴趣单元更新与检索过程相结合，IRA 能够持续适应不断变化的用户偏好，确保稳健且精细的个性化，而不受过去训练分布的限制。 我们通过在现实世界数据集上的大量实验验证了 IRA 的有效性，包括将其部署在韩国领先社区平台 NAVER 的 CAFE 的首页部分。