AutoJourn: Multi-Perspective Summarisation, Bias Detection and Bias Neutralisation for LLM-Generated News in Automated Journalism

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Himel Ghosh, Ahmed Mosharafa, Georg Groh
• Venue: arXiv: cs.CL
• Published: 2026-07-21
• Collected: 2026-07-22
• URL: http://arxiv.org/abs/2607.18983v1
• PDF: https://arxiv.org/pdf/2607.18983v1
• DOI: UNKNOWN

Plain-Language Brief

The researchers present AutoJourn, a system for generating news articles using large language models (LLMs) that consider multiple perspectives and aim to reduce bias in AI-generated news. The system includes components for extracting diverse viewpoints from social media discussions, creating balanced summaries, and detecting/mitigating bias.

What the Document Says

• The researchers present AutoJourn, a system for news generation using LLMs.
• The system addresses three challenges in automated journalism: perspective extraction, summary preservation of viewpoint diversity, and bias detection/mitigation.
• The pipeline includes advanced prompt engineering, retrieval augmentation, multi-perspective summarisation module, and a bias analysis suite.
• The components produce diverse perspective sets, balanced summaries, and support sentence-level bias detection, type classification, and neutralization.
• Users can inspect perspective clusters, compare stance-specific summaries, generate news articles, and apply bias-aware rewrites directly in the interface.
• Evaluation of each component is done with intrinsic metrics such as semantic diversity, summary quality, and bias reduction. Improvements over strong baselines are shown while maintaining content fidelity.
• A live demo accompanies the paper for further research on socially responsible automated journalism.

Key Concepts Mentioned

• AutoJourn
• large language models (LLMs)
• multi-perspective news generation
• bias detection
• bias mitigation
• summarization
• social media discussions

Claims Made by the Document

• Improvements over strong baselines are shown in the evaluation.
• The system maintains content fidelity.

Evidence Used

• Users can inspect perspective clusters, compare stance-specific summaries, generate news articles, and apply bias-aware rewrites directly in the interface.
• Evaluation of each component is done with intrinsic metrics such as semantic diversity, summary quality, and bias reduction.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

We present AutoJourn, a demonstration system for multi-perspective news generation and bias-aware evaluation using large language models (LLMs). The system tackles three core challenges in responsible automated journalism: extracting diverse perspectives from unstructured social media discussions, generating summaries that preserve viewpoint diversity, and detecting or mitigating bias in AI-generated news. The pipeline integrates advanced prompt engineering with optional retrieval augmentation to produce semantically diverse perspective sets, a multi-perspective summarisation module that merges conflicting viewpoints into balanced summaries, and a bias analysis suite supporting sentence-level bias detection and type classification in the generated news article, and automatic neutralisation. Users can inspect perspective clusters, compare stance-specific summaries, generate news articles, and apply bias-aware rewrites directly in the interface. We evaluate each component with intrinsic metrics -- semantic diversity, summary quality, and bias reduction and show improvements over strong baselines while maintaining content fidelity. A live, publicly accessible demo accompanies the paper to facilitate reproducibility and further research on socially responsible automated journalism.