Adding a Neutral Third Agent to LLM Deliberation: An Exploratory Case Series of Six Dyadic and Triadic Conversations

Nov 4, 2026·
Abbas Hamidavi
Abbas Hamidavi
· 0 min read
Abstract
When two large language models deliberate toward a joint answer, does adding a third, neutral agent change what the pair produces? This study compares matched dyadic and triadic deliberations in two output formats, free discussion and structured revision, across six deliberations. Transcripts, the supplementary prompt set, and the coding materials are openly deposited.
Type
Publication
Ethics and Information Technology (under review)
publications
Large Language Models Machine Psychology Deliberation Multi-Agent Systems
Abbas Hamidavi
Authors
Abbas Hamidavi (he/him)
CS Student & AI Researcher Open Science advocate Building tools to probe, analyze, and secure LLM architectures Exploring the intersection of Red Teaming and Interpretability. AI Safety Researcher at Islamic Azad University, Mashhad, running an independent research programme in Machine Psychology. ORCID: https://orcid.org/0009-0001-7342-7844
α