Search papers, labs, and topics across Lattice.
This paper introduces Multi-Agent LLM-based Emotional deToxification (MALLET), a system designed to mitigate emotional stimulation from online content using a multi-agent architecture. MALLET employs a BERT-based emotion classifier to quantify stimulus intensity and an LLM to rewrite texts into BALANCED and COOL presentation modes, while monitoring consumption patterns to provide personalized recommendations. Experiments on the AG News dataset demonstrated a significant reduction in stimulus scores (up to 19.3%) and improved emotion balance, with category-level analysis showing substantial reductions in Sports, Business, and Sci/Tech news.
LLMs can be used to detoxify emotionally stimulating news content, reducing its intensity by up to 33% in certain categories without sacrificing semantic meaning.
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.
