Saved in:
Bibliographic Details
Main Author: Marandi, Ramtin Zargari
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.28813
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912989765959680
author Marandi, Ramtin Zargari
author_facet Marandi, Ramtin Zargari
contents In multi-agent debate (MAD) systems, performance gains are often reported; however, because the debate protocol (e.g., number of agents, rounds, and aggregation rule) is typically held fixed while model-related factors vary, it is difficult to disentangle protocol effects from model effects. To isolate these effects, we compare three main protocols, Within-Round (WR; agents see only current-round contributions), Cross-Round (CR; full prior-round context), and novel Rank-Adaptive Cross-Round (RA-CR; dynamically reorders agents and silences one per round via an external judge model), against a No-Interaction baseline (NI; independent responses without peer visibility). In a controlled macroeconomic case study (20 diverse events, five random seeds, matched prompts/decoding), RA-CR achieves faster convergence than CR, WR shows higher peer-referencing, and NI maximizes Argument Diversity (unaffected across the main protocols). These results reveal a trade-off between interaction (peer-referencing rate) and convergence (consensus formation), confirming protocol design matters. When consensus is prioritized, RA-CR outperforms the others.
format Preprint
id arxiv_https___arxiv_org_abs_2603_28813
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle The impact of multi-agent debate protocols on debate quality: a controlled case study
Marandi, Ramtin Zargari
Multiagent Systems
Artificial Intelligence
In multi-agent debate (MAD) systems, performance gains are often reported; however, because the debate protocol (e.g., number of agents, rounds, and aggregation rule) is typically held fixed while model-related factors vary, it is difficult to disentangle protocol effects from model effects. To isolate these effects, we compare three main protocols, Within-Round (WR; agents see only current-round contributions), Cross-Round (CR; full prior-round context), and novel Rank-Adaptive Cross-Round (RA-CR; dynamically reorders agents and silences one per round via an external judge model), against a No-Interaction baseline (NI; independent responses without peer visibility). In a controlled macroeconomic case study (20 diverse events, five random seeds, matched prompts/decoding), RA-CR achieves faster convergence than CR, WR shows higher peer-referencing, and NI maximizes Argument Diversity (unaffected across the main protocols). These results reveal a trade-off between interaction (peer-referencing rate) and convergence (consensus formation), confirming protocol design matters. When consensus is prioritized, RA-CR outperforms the others.
title The impact of multi-agent debate protocols on debate quality: a controlled case study
topic Multiagent Systems
Artificial Intelligence
url https://arxiv.org/abs/2603.28813