FUM: Fine-grained and Fast User Modeling for News Recommendation

TL;DR

This paper introduces FUM, a novel user modeling framework that captures detailed word-level interactions across clicked news to improve news recommendation accuracy and efficiency.

Contribution

FUM transforms user modeling into a document modeling task using an efficient transformer, capturing intra- and inter-news interactions for better recommendations.

Findings

FUM outperforms existing methods on real-world datasets.

FUM effectively models fine-grained user interests.

FUM is computationally efficient for long documents.

Abstract

User modeling is important for news recommendation. Existing methods usually first encode user's clicked news into news embeddings independently and then aggregate them into user embedding. However, the word-level interactions across different clicked news from the same user, which contain rich detailed clues to infer user interest, are ignored by these methods. In this paper, we propose a fine-grained and fast user modeling framework (FUM) to model user interest from fine-grained behavior interactions for news recommendation. The core idea of FUM is to concatenate the clicked news into a long document and transform user modeling into a document modeling task with both intra-news and inter-news word-level interactions. Since vanilla transformer cannot efficiently handle long document, we apply an efficient transformer named Fastformer to model fine-grained behavior interactions. Extensive experiments on two real-world datasets verify that FUM can effectively and efficiently model user interest for news recommendation.