A multi-agent LLM chatbot with separate safety, HR, and peer personas outperformed a single generic chatbot on usability, psychological needs, social presence, and trust in a 12-person role-play study.
Hey Alice, what do you think?
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.HC 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Supporting Construction Worker Well-Being with a Multi-Agent Conversational AI System
A multi-agent LLM chatbot with separate safety, HR, and peer personas outperformed a single generic chatbot on usability, psychological needs, social presence, and trust in a 12-person role-play study.