Where Reasoning Diverges: Localized Multi-Agent Debate

Abstract:Multi-agent debate commonly exchanges complete reasoning traces even when disagreements concern only a few intermediate claims. We introduce Localized Multi-Agent Debate (LMAD), an inference-time protocol that represents agent traces as typed nodes, locates their earliest conflict, and restricts debate to the corresponding local segments. Guarded resolution extends a shared committed state so that later conflicts can be addressed without reopening accepted steps. We evaluate a single fixed LMAD configuration on four multi-hop question-answering benchmarks using ten backbones from four model families. Our method achieves the highest macro-averaged judge accuracy across all ten backbones, outperforming the strongest conventional baseline by up to 7.20 percentage points.

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

arXiv:2608.01463 [cs.AI]

(or arXiv:2608.01463v1 [cs.AI] for this version)

https://doi.org/10.48550/arXiv.2608.01463 Focus to learn more arXiv-issued DOI via DataCite (pending registration)

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.