Panoramic Interests: Stylistic-Content Aware Personalized Headline Generation
This work improves personalization for news readers by incorporating stylistic preferences, though it is incremental as it builds on existing user-oriented methods.
The paper tackles personalized news headline generation by addressing the oversight of stylistic preferences in existing methods, proposing the SCAPE framework that integrates both content and stylistic features using LLM collaboration and contrastive learning, achieving superior performance on the PENS dataset.
Personalized news headline generation aims to provide users with attention-grabbing headlines that are tailored to their preferences. Prevailing methods focus on user-oriented content preferences, but most of them overlook the fact that diverse stylistic preferences are integral to users' panoramic interests, leading to suboptimal personalization. In view of this, we propose a novel Stylistic-Content Aware Personalized Headline Generation (SCAPE) framework. SCAPE extracts both content and stylistic features from headlines with the aid of large language model (LLM) collaboration. It further adaptively integrates users' long- and short-term interests through a contrastive learning-based hierarchical fusion network. By incorporating the panoramic interests into the headline generator, SCAPE reflects users' stylistic-content preferences during the generation process. Extensive experiments on the real-world dataset PENS demonstrate the superiority of SCAPE over baselines.