Reaching Agreement Among Reasoning LLM Agents

TL;DR

This paper introduces Aegean, a formal consensus protocol for multi-agent reasoning systems that improves efficiency and reliability by reducing latency and ensuring correctness in collective AI reasoning tasks.

Contribution

It presents a formal model for multi-agent refinement, proposes the Aegean consensus protocol, and demonstrates significant latency reductions with provable guarantees.

Findings

Reduces reasoning latency by up to 20 times.

Maintains answer quality within 2.5%.

Provides formal safety and liveness guarantees.

Abstract

Multi-agent systems have extended the capability of agentic AI. Instead of single inference passes, multiple agents perform collective reasoning to derive high quality answers. However, existing multi-agent orchestration relies on static heuristic workflows such as fixed loop limits and barrier synchronization. These ad-hoc approaches waste computational resources, incur high latency due to stragglers, and risk finalizing transient agreements. We argue that reliable multi-agent reasoning requires a formal foundation analogous to classical distributed consensus problem. To that end, we propose a formal model of the multi-agent refinement problem. The model includes definitions of the correctness guarantees and formal semantics of agent reasoning. We then introduce Aegean, a consensus protocol designed for stochastic reasoning agents that solves multi-agent refinement. We implement the protocol in Aegean-Serve, a consensus-aware serving engine that performs incremental quorum detection across concurrent agent executions, enabling early termination when sufficient agents converge. Evaluation using four mathematical reasoning benchmarks shows that Aegean provides provable safety and liveness guarantees while reducing latency by 1.2--20$\times$ compared to state-of-the-art baselines, maintaining answer quality within 2.5%. Consistent gains across both local GPU deployments and commercial API providers validate that consensus-based orchestration eliminates straggler delays without sacrificing correctness.