A Hashgraph-Inspired Consensus Mechanism for Reliable Multi-Model Reasoning

TL;DR

The paper tackles the problem of inconsistent and hallucinatory outputs from multiple black-box reasoning models by introducing a Hashgraph-inspired consensus mechanism that treats each model as a node in a distributed ledger-like network. Through gossip-about-gossip information exchange and virtual voting, the ensemble iteratively updates outputs $O_i^r$ to converge on a final consensus $O^*$, tolerating Byzantine-like faults up to $1/3$. It contributes a formal design for the consensus process, a prototype architecture with modules for prompting, evaluation, and data flow, and an evaluation framework focused on factual accuracy, hallucination reduction, convergence rounds, and robustness. The approach aims to provide a decentralized, self-validating AI system that preserves useful minority contributions while pruning nonfactual content, potentially enabling more reliable multi-vendor AI collaborations in high-stakes tasks.

Abstract

Inconsistent outputs and hallucinations from large language models (LLMs) are major obstacles to reliable AI systems. When different proprietary reasoning models (RMs), such as those by OpenAI, Google, Anthropic, DeepSeek, and xAI, are given the same complex request, they often produce divergent results due to variations in training and inference. This paper proposes a novel consensus mechanism, inspired by distributed ledger technology, to validate and converge these outputs, treating each RM as a black-box peer. Building on the Hashgraph consensus algorithm, our approach employs gossip-about-gossip communication and virtual voting to achieve agreement among an ensemble of RMs. We present an architectural design for a prototype system in which RMs iteratively exchange and update their answers, using information from each round to improve accuracy and confidence in subsequent rounds. This approach goes beyond simple majority voting by incorporating the knowledge and cross-verification content of every model. We justify the feasibility of this Hashgraph-inspired consensus for AI ensembles and outline its advantages over traditional ensembling techniques in reducing nonfactual outputs. Preliminary considerations for implementation, evaluation criteria for convergence and accuracy, and potential challenges are discussed. The proposed mechanism demonstrates a promising direction for multi-agent AI systems to self-validate and deliver high-fidelity responses in complex tasks.