Council Mode: A Heterogeneous Multi-Agent Consensus Framework for Reducing LLM Hallucination and Bias

TL;DR

The paper introduces the Council Mode, a multi-agent consensus framework that reduces hallucinations and biases in LLMs by synthesizing outputs from diverse models, significantly improving factual accuracy and reliability.

Contribution

It presents a novel multi-agent consensus approach that dispatches queries to heterogeneous LLMs and synthesizes their outputs, reducing hallucinations and biases compared to individual models.

Findings

35.9% reduction in hallucination rates on HaluEval

7.8-point improvement on TruthfulQA

10.2-point improvement on MDR-500 benchmark

Abstract

Large Language Models (LLMs) have demonstrated advanced capabilities but often suffer from factual inaccuracies (hallucinations) and systematic biases. These issues, sometimes amplified in specific architectures like Mixture-of-Experts (MoE) which motivate our work, pose risks for reliable deployment. To address these challenges, we propose the Council Mode, a multi-agent consensus framework. Our approach dispatches queries to multiple heterogeneous frontier LLMs in parallel and synthesizes their outputs using a dedicated consensus model. The pipeline consists of three phases: an intelligent triage for query complexity, parallel generation across diverse models, and a structured synthesis that identifies agreement, disagreement, and unique findings. In our evaluation, conducted under controlled no-web settings, the Council Mode achieved a 35.9% relative reduction in hallucination rates on a 1,200-sample HaluEval subset and a 7.8-point improvement on TruthfulQA compared to the top-performing individual model. On our curated MDR-500 multi-domain reasoning benchmark, the Council Mode achieved a Quality Score of 91.7%, representing a 10.2-point improvement over the best individual model. The framework also exhibited lower measured bias variance under our rubric-based evaluation protocol. We provide a cost-effectiveness analysis showing that the framework incurs a 4.2x token-cost overhead, making it most suitable for accuracy-prioritized applications where the cost of errors exceeds the added inference cost. These findings suggest that structured multi-agent consensus is a promising direction for enhancing the reliability and factual grounding of LLM-generated content.