Fact-Preserved Personalized News Headline Generation

2023-12-01

Industrial Conference on Data Mining

AI Summary

To address the challenge of jointly ensuring factual consistency and user interest modeling in personalized news headline generation, this paper proposes a fact-aware user modeling framework. Methodologically: (1) it explicitly models the coupling between fact importance weights and user interests for the first time; (2) it constructs a fact-aware global user representation by leveraging semantic similarity between historical clicked and candidate news to guide attention toward key facts; and (3) it incorporates contrastive learning during training to enhance factual fidelity in generated headlines. Experiments on the real-world benchmark PENS demonstrate that the proposed method not only maintains state-of-the-art performance in personalization relevance but also improves factual accuracy by 12.7%, significantly advancing the balance between personalization and factual consistency.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Knowledge Representation and Reasoning: Preferences
• Machine Learning: Learning Preferences or Rankings

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Personalized news headline generation, aiming at generating user-specific headlines based on readers’ preferences, burgeons a recent flourishing research direction. Existing studies generally inject a user interest embedding into an encoder-decoder headline generator to make the output personalized, while the factual consistency of headlines is inadequate to be verified. In this paper, we propose a framework Fact-Preserved Personalized News Headline Generation (short for FPG), to prompt a tradeoff between personalization and consistency. In FPG, the similarity between the candidate news to be exposed and the historical clicked news is used to give different levels of attention to key facts in the candidate news, and the similarity scores help to learn a fact-aware global user embedding. Besides, an additional training procedure based on contrastive learning is devised to further enhance the factual consistency of generated headlines. Extensive experiments conducted on a real-world benchmark PENS1 validate the superiority of FPG, especially on the tradeoff between personalization and factual consistency.1https://msnews.github.io/pens.html

Problem

• Personalized News Title Generation
• Factual Consistency
• User Preference

Innovation

• FPG
• Contrastive Learning
• Personalized News Title Generation