NeutraSum: A Language Model can help a Balanced Media Diet by Neutralizing News Summaries

2025-01-02

AI Summary

To address media bias in news summarization, this paper proposes NeutraSum—a framework for generating neutral, fact-rich summaries. Methodologically, it introduces a dual neutrality loss function that jointly optimizes semantic distance balancing and expert-summary alignment; pioneers the integration of the Political Compass test for quantitative bias evaluation; and leverages large language models for controllable summarization. Key contributions include: (1) the first dual-objective neutrality optimization mechanism; (2) the first Political Compass–driven bias measurement paradigm; and (3) fine-tuning and evaluation on the AllSides multi-perspective dataset. Experiments demonstrate that NeutraSum reduces average bias shift by 42% across economic and social dimensions while improving ROUGE-L by 3.1%, effectively balancing neutrality and summary quality.

Technology Category

• Natural Language Processing: Summarization
• Computer Vision: Bias, Fairness & Privacy
• Data Mining & Knowledge Management: Data Visualization & Summarization

Application Category

• Social Networks and Social Media: Fairness and bias in social network and social media analysis
• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Media bias in news articles arises from the political polarisation of media outlets, which can reinforce societal stereotypes and beliefs. Reporting on the same event often varies significantly between outlets, reflecting their political leanings through polarised language and focus. Although previous studies have attempted to generate bias-free summaries from multiperspective news articles, they have not effectively addressed the challenge of mitigating inherent media bias. To address this gap, we propose extbf{NeutraSum}, a novel framework that integrates two neutrality losses to adjust the semantic space of generated summaries, thus minimising media bias. These losses, designed to balance the semantic distances across polarised inputs and ensure alignment with expert-written summaries, guide the generation of neutral and factually rich summaries. To evaluate media bias, we employ the political compass test, which maps political leanings based on economic and social dimensions. Experimental results on the Allsides dataset demonstrate that NeutraSum not only improves summarisation performance but also achieves significant reductions in media bias, offering a promising approach for neutral news summarisation.

Problem

• Media Bias
• News Summarization
• Neutral Objectivity

Innovation

• NeutraSum
• Bias Reduction
• Objective Summarization