Analyst News model announce report study agent arxiv

The impact of multi-agent debate protocols on debate quality: a controlled case study

arXiv cs.MAby Ramtin Zargari MarandiApril 1, 20261 min read0 views

arXiv:2603.28813v1 Announce Type: new Abstract: 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 see

View PDF HTML (experimental)

Abstract: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.

Comments: 16 pages, 3 figures

Subjects:

Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI)

Cite as: arXiv:2603.28813 [cs.MA]

(or arXiv:2603.28813v1 [cs.MA] for this version)

https://doi.org/10.48550/arXiv.2603.28813

arXiv-issued DOI via DataCite (pending registration)