JRE-L: Journalist, Reader, and Editor LLMs in the Loop for Science Journalism for the General Audience

2025-01-28

AI Summary

Scientific news often suffers from low public accessibility due to technical jargon and conceptual complexity. To address this, we propose a novel three-role large language model (LLM) collaboration framework—comprising simulated *Journalist*, *Reader*, and *Editor* agents—that explicitly incorporates public comprehension feedback into the generation pipeline, establishing a closed-loop “write–read–feedback–revise” process. Methodologically, we adopt a lightweight multi-LLM orchestration paradigm: three open-weight models (two 7B and one 1.8B variants) are coordinated via role-specific prompt engineering, augmented by an iterative, feedback-driven rewriting mechanism and a reproducible evaluation protocol. Experiments demonstrate that our approach significantly outperforms GPT-4 and existing LLM collaboration methods across readability, factual accuracy, and public friendliness. The implementation and evaluation framework are publicly released.

Technology Category

• Natural Language Processing: (Large) Language Models
• Machine Learning: Large Multimodal Models (LMMs)
• Humans and AI: AI for Accessibility

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Large language models for search

Abstract

Science journalism reports current scientific discoveries to non-specialists, aiming to enable public comprehension of the state of the art. This task is challenging as the audience often lacks specific knowledge about the presented research. We propose a JRE-L framework that integrates three LLMs mimicking the writing-reading-feedback-revision loop. In JRE-L, one LLM acts as the journalist, another LLM as the general public reader, and the third LLM as an editor. The journalist's writing is iteratively refined by feedback from the reader and suggestions from the editor. Our experiments demonstrate that by leveraging the collaboration of two 7B and one 1.8B open-source LLMs, we can generate articles that are more accessible than those generated by existing methods, including prompting single advanced models such as GPT-4 and other LLM-collaboration strategies. Our code is publicly available at github.com/Zzoay/JRE-L.

Problem

• science communication
• complex concepts
• public understanding

Innovation

• JRE-L
• Cyclical Collaborative Framework
• Enhanced Readability