Scholarlink

In today’s hyperconnected world, news sources are widely accessible, yet challenges persist in managing redundancy, retrieval efficiency, and content personalization. This paper presents a modular automated newsroom leveraging Large Language Models (LLMs) to streamline news processing and enhance editorial workflows. The system employs a structured pipeline of LLM-powered agents, each performing specialized tasks in sequence to transform raw news data from multiple sources into enriched, structured content. This content is stored in a database, making it accessible via API-driven services for editorial applications. By integrating Retrieval-Augmented Generation (RAG), the framework enables semantic search, intelligent content retrieval, and real-time editorial automation, enhancing discoverability and efficiency within a scalable, Service-Oriented Architecture.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Automated Newsrooms and Enhanced Editorial Processes Through Large Language Models

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Dario Pellegrini,
  • Davide Ragazzi

摘要

In today’s hyperconnected world, news sources are widely accessible, yet challenges persist in managing redundancy, retrieval efficiency, and content personalization. This paper presents a modular automated newsroom leveraging Large Language Models (LLMs) to streamline news processing and enhance editorial workflows. The system employs a structured pipeline of LLM-powered agents, each performing specialized tasks in sequence to transform raw news data from multiple sources into enriched, structured content. This content is stored in a database, making it accessible via API-driven services for editorial applications. By integrating Retrieval-Augmented Generation (RAG), the framework enables semantic search, intelligent content retrieval, and real-time editorial automation, enhancing discoverability and efficiency within a scalable, Service-Oriented Architecture.