Courtroom-Style Multi-Agent Debate with Progressive RAG and Role-Switching for Controversial Claim Verification

TL;DR

PROClaim introduces a structured courtroom-style multi-agent debate framework with Progressive RAG and role-switching, significantly improving the reliability of claim verification in high-stakes scenarios.

Contribution

It presents a novel structured debate framework with dynamic evidence refinement and model heterogeneity, outperforming existing methods on the Check-COVID benchmark.

Findings

Achieves 81.7% accuracy on Check-COVID, 10% higher than standard debate.

P-RAG contributes +7.5 percentage points to performance.

Structural deliberation reduces systematic biases.

Abstract

Large language models (LLMs) remain unreliable for high-stakes claim verification due to hallucinations and shallow reasoning. While retrieval-augmented generation (RAG) and multi-agent debate (MAD) address this, they are limited by one-pass retrieval and unstructured debate dynamics. We propose a courtroom-style multi-agent framework, PROClaim, that reformulates verification as a structured, adversarial deliberation. Our approach integrates specialized roles (e.g., Plaintiff, Defense, Judge) with Progressive RAG (P-RAG) to dynamically expand and refine the evidence pool during the debate. Furthermore, we employ evidence negotiation, self-reflection, and heterogeneous multi-judge aggregation to enforce calibration, robustness, and diversity. In zero-shot evaluations on the Check-COVID benchmark, PROClaim achieves 81.7% accuracy, outperforming standard multi-agent debate by 10.0 percentage points, with P-RAG driving the primary performance gains (+7.5 pp). We ultimately demonstrate that structural deliberation and model heterogeneity effectively mitigate systematic biases, providing a robust foundation for reliable claim verification. Our code and data are publicly available at https://github.com/mnc13/PROClaim.