L-MAD: A Systematic Evaluation of Multi-Agent Debate Structures in Legal Reasoning

Researchers introduced L-MAD, a framework for evaluating multi-agent debate structures in legal reasoning, finding that increasing agent population improves accuracy but extending discussion rounds causes over-deliberation drift where agents reinforce errors. The system outperformed single-agent baselines by up to 8% in legal textual entailment tasks.

arXiv:2607.09099v1 Announce Type: new Abstract: While multi-agent debate MAD frameworks have shown significant potential in general reasoning, their effectiveness in highly structured, knowledge-heavy legal domains remains under-explored. In this work, we introduce the Legal Multi-Agent Debate L-MAD framework to systematically evaluate different debate structures and aggregation methods within Legal Textual Entailment. By assigning distinct expert personas to multiple agents, L-MAD improves upon strong single-agent baselines by up to 8\%. Furthermore, analyzing how debate scales reveals a clear trade-off: increasing the agent population reduces inconsistency and improves accuracy, whereas extending discussion rounds induces a detrimental \textit{over-deliberation drift} where agents reinforce each other's mistakes. Ultimately, our findings outline the practical boundaries and safety margins of deploying collaborative multi-agent systems in high-stakes legal reasoning environments.