Efficient Pointwise-Pairwise Learning-to-Rank for News Recommendation
This work addresses the problem of scalable and effective news recommendation for users, representing an incremental improvement by combining existing approaches with theoretical guarantees.
The paper tackles the challenge of news recommendation by integrating pointwise and pairwise learning-to-rank methods to balance computational efficiency and ranking accuracy, achieving state-of-the-art performance on MIND and Adressa datasets.
News recommendation is a challenging task that involves personalization based on the interaction history and preferences of each user. Recent works have leveraged the power of pretrained language models (PLMs) to directly rank news items by using inference approaches that predominately fall into three categories: pointwise, pairwise, and listwise learning-to-rank. While pointwise methods offer linear inference complexity, they fail to capture crucial comparative information between items that is more effective for ranking tasks. Conversely, pairwise and listwise approaches excel at incorporating these comparisons but suffer from practical limitations: pairwise approaches are either computationally expensive or lack theoretical guarantees, and listwise methods often perform poorly in practice. In this paper, we propose a novel framework for PLM-based news recommendation that integrates both pointwise relevance prediction and pairwise comparisons in a scalable manner. We present a rigorous theoretical analysis of our framework, establishing conditions under which our approach guarantees improved performance. Extensive experiments show that our approach outperforms the state-of-the-art methods on the MIND and Adressa news recommendation datasets.