Unlike traditional single-model approaches, our system implements an ensemble of specialized medical "expert" agents, each represented by an individual LLM, mimicking real-world clinical triage and decision-making.

Despite the growing clinical adoption of large language models (LLMs), current approaches heavily rely on single model architectures. To overcome risks of obsolescence and rigid dependence on single model systems, we present a novel framework, termed the Consensus Mechanism. Mimicking clinical triage and multidisciplinary clinical decision-making, the Consensus Mechanism implements an ensemble of specialized medical expert agents, enabling improved clinical decision-making while maintaining robust adaptability. This architecture enables the Consensus Mechanism to be optimized for cost, latency, or performance, purely based on its internal model configuration. Tryout: consensus.sully.ai here.

how does SullyAI manage disagreement between the individual expert agents? Do you treat some agents as weighted more heavily or is it majority rules?

@aisha_kabir Aisha, see the paper here https://arxiv.org/abs/2505.23075 of how we use a weighted log-opinion pool then apply cascade boosting to the confidence score/probabilities

Can I see something without logging in?

@admiralrohan Yes visit https://www.sully.ai/ @ahmedomar The link to the website links directly to the signup / signin rather than the intro landing page.

Sully.ai

SullyAI: 10x clinical diagnosis - introducing Consensus