L-MAD: A Systematic Evaluation of Multi-Agent Debate Structures in Legal Reasoning
This work provides practical guidelines for deploying multi-agent systems in high-stakes legal reasoning, highlighting trade-offs between agent count and discussion rounds.
L-MAD systematically evaluates multi-agent debate structures in legal reasoning, achieving up to 8% improvement over single-agent baselines, while revealing that increasing agents reduces inconsistency but more rounds cause over-deliberation drift.
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.