PCRec: A Multi-Interest News Recommendation Framework with Prompt-Guided Cross-view Contrastive Learning

Effective news article recommendation is crucial for alleviating users' information overload. While recent prompt-based news recommendation methods have shown promising performance by reformulating the recommendation task as a masked prediction problem, this paradigm still faces several major limitations: a) Inadequate multi-interest representation, existing approaches typically transform users' historical clicks into a prompt template and represent user preferences through a single embedding vector. This oversimplified representation fails to capture the inherent diversity of user interests; b) Limited global interaction modeling, current methods primarily focus on local user-news interactions, failing to leverage the rich contextual information embedded in global inter-user and inter-news distributions; c) Historical interaction truncation, the input length constraints of language models force current approaches to truncate or compress users' historical interactions, which is particularly problematic for news recommendation given the typically extensive length of articles. To address these problems, this paper proposes PCRec, a prompt-guided cross-view contrastive learning framework for multi-interest news recommendation. PCRec first introduces feature-level prompts to overcome the input constraints inherent in text-level prompts. Moreover, a two-stage user modeling module is designed to capture users' multi-interests. Finally, to model global user-news relationships, PCRec implements a cross-view contrastive learning strategy. This approach groups similar users, enabling learning from multiple perspectives and breaking down isolated relationships among users, news categories, and news subcategories. Extensive experiments on two real-world news recommendation datasets validate the superiority of our proposed PCRec compared with various state-of-the-art news recommendation baselines. Our code is available at https://github.com/zgzjdx/PCRec.