Multi-Agent Large Language Model Based Emotional Detoxification Through Personalized Intensity Control for Consumer Protection

2026-02-26

AI Summary

In the attention economy, sensationalist news content often triggers excessive emotional arousal, impairing users’ rational decision-making. To address this, this work proposes MALLET, a multi-agent affective detoxification system that integrates four specialized agents—emotion analysis, regulation, balance monitoring, and personalized guidance—to quantify emotional intensity in news texts and generate neutralized rewrites tailored to individual user sensitivity. The system innovatively decouples control over emotional stimulation intensity from semantic fidelity, introducing a dual-mode neutral presentation (BALANCED/COOL) alongside a personalized emotion regulation mechanism. Evaluated on 800 AG News articles, MALLET reduces average emotional arousal scores by 19.3% (ranging from 17.8% to 33.8% across sports, business, and technology categories), significantly enhancing emotional balance while preserving semantic consistency.

Technology Category

• Cognitive Modeling & Cognitive Systems: Affective Computing
• Multiagent Systems: Agreement, Argumentation & Negotiation
• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Economics, Online Markets and Human Computation: Fairness, privacy, and diversity in economic environments

Abstract

In the attention economy, sensational content exposes consumers to excessive emotional stimulation, hindering calm decision-making. This study proposes Multi-Agent LLM-based Emotional deToxification (MALLET), a multi-agent information sanitization system consisting of four agents: Emotion Analysis, Emotion Adjustment, Balance Monitoring, and Personal Guide. The Emotion Analysis Agent quantifies stimulus intensity using a 6-emotion BERT classifier, and the Emotion Adjustment Agent rewrites texts into two presentation modes, BALANCED (neutralized text) and COOL (neutralized text + supplementary text), using an LLM. The Balance Monitoring Agent aggregates weekly information consumption patterns and generates personalized advice, while the Personal Guide Agent recommends a presentation mode according to consumer sensitivity. Experiments on 800 AG News articles demonstrated significant stimulus score reduction (up to 19.3%) and improved emotion balance while maintaining semantic preservation. Near-zero correlation between stimulus reduction and semantic preservation confirmed that the two are independently controllable. Category-level analysis revealed substantial reduction (17.8-33.8%) in Sports, Business, and Sci/Tech, whereas the effect was limited in the World category, where facts themselves are inherently high-stimulus. The proposed system provides a framework for supporting calm information reception of consumers without restricting access to the original text.

Problem

• Emotional Detoxification
• Consumer Protection
• Attention Economy
• Emotion Regulation
• Information Sanitization

Innovation

• Multi-Agent LLM
• Emotional Detoxification
• Personalized Intensity Control
• Stimulus-Neutralized Rewriting
• Semantic Preservation