Search papers, labs, and topics across Lattice.
The paper introduces Case-Adaptive Multi-agent Panel (CAMP), a novel framework for clinical prediction that addresses case-level heterogeneity in LLM outputs by dynamically assembling specialist panels based on diagnostic uncertainty. CAMP employs three-valued voting (KEEP/REFUSE/NEUTRAL) to allow specialists to abstain outside their expertise, and uses a hybrid router to combine specialist consensus, attending physician judgment, and evidence-based arbitration. Experiments on MIMIC-IV using various LLMs demonstrate that CAMP outperforms strong baselines while using fewer tokens and providing transparent decision audits.
LLMs can make better clinical predictions by dynamically assembling expert panels tailored to the complexity of each case, rather than relying on fixed roles and simple majority voting.
Large language models applied to clinical prediction exhibit case-level heterogeneity: simple cases yield consistent outputs, while complex cases produce divergent predictions under minor prompt changes. Existing single-agent strategies sample from one role-conditioned distribution, and multi-agent frameworks use fixed roles with flat majority voting, discarding the diagnostic signal in disagreement. We propose CAMP (Case-Adaptive Multi-agent Panel), where an attending-physician agent dynamically assembles a specialist panel tailored to each case's diagnostic uncertainty. Each specialist evaluates candidates via three-valued voting (KEEP/REFUSE/NEUTRAL), enabling principled abstention outside one's expertise. A hybrid router directs each diagnosis through strong consensus, fallback to the attending physician's judgment, or evidence-based arbitration that weighs argument quality over vote counts. On diagnostic prediction and brief hospital course generation from MIMIC-IV across four LLM backbones, CAMP consistently outperforms strong baselines while consuming fewer tokens than most competing multi-agent methods, with voting records and arbitration traces offering transparent decision audits.
