Multi-Agent Debate (MAD) improves LLM-agent accuracy but suffers from rapid context growth, limiting scalability in larger multi-agent settings. Existing methods prune low-utility communications using prior signals, such as token-level log-likelihoods or LLM self-reported confidence. However, these signals become unreliable under hallucination, degrading the accuracy of MAD methods that rely on them. We propose SVR-MAD, a Bayesian-inspired MAD framework that treats pre-debate signals as priors and debate outcomes as posterior-style evidence for estimating agent correctness. SVR-MAD uses this evidence to incrementally construct the communication graph, prioritizing agents whose answers survive peer challenges. Experiments across multiple LLMs and benchmarks show that SVR-MAD reduces token cost by up to 61% while matching or improving accuracy relative to the most accurate competing MAD baseline.
@article{arxiv.2605.23099,
title = {SVR-MAD: A Bayesian-Inspired Framework for Posterior-Guided Multi-Agent Debate},
author = {Weifan Jiang and Rana Shahout and Minghao Li and Zhenting Qi and Yilun Du and Michael Mitzenmacher and Minlan Yu},
journal= {arXiv preprint arXiv:2605.23099},
year = {2026}
}