{"id":"doi:10.64898/2026.02.22.26346818","kind":"preprint","title":"Agent Role Structure and Operating Characteristics in Large Language Model Clinical Classification: A Comparative Study of Specialist and Deliberative Multi-Agent Protocols","authors":["Anderson C"],"author_count":1,"year":2026,"published":"2026-02-24","venue":"medRxiv","cited_by":0,"doi":"10.64898/2026.02.22.26346818","arxiv_id":null,"nct_id":null,"source_url":"https://doi.org/10.64898/2026.02.22.26346818","archive_url":"https://space-kitty.com/papers/doi:10.64898/2026.02.22.26346818","updated":null,"pmid":null,"pmcid":null,"openalex_id":null,"pdf_url":null,"abstract":"Large language models (LLMs) are increasingly considered in clinical decision support, yet the architectural effects of role decomposition within multi-agent systems remain poorly isolated. Prior comparisons of single-agent and multi-agent prompting often confound workflow structure with changes in model configuration, training, or decoding. We present a controlled study evaluating whether specialist and generalist multi-agent protocols produce different operating characteristics under fixed inference conditions. Two deterministic protocols, Generic Deliberative (GD) and Feature-Specialist (FS), are evaluated under identical base weights, decoding settings, computational budget, and adjudication logic. The primary comparison isolates role structure and information routing, while feature-assignment and model-scale ablations evaluate boundary conditions. On the UCI Cleveland Heart Disease and Pima Indians Diabetes benchmarks, altering role structure produces substantial operating-characteristic shifts. Under the primary configuration ( llama3.1:8b ), Cleveland FS achieves higher accuracy (0.72 vs. 0.65) and macro-F1 (0.71 vs. 0.65), with a specificity-favoring operating point, whereas Pima GD achieves higher accuracy (0.68 vs. 0.51) and macro-F1 (0.64 vs. 0.49), while FS induces pronounced class asymmetry. Feature-pair ablations show that specialist feature assignment is a design lever, with FS accuracy ranging from 0.41 to 0.72. A model-scale ablation ( qwen2.5:14b ) shows that protocol-level effects interact with base-model capacity. Overall, internal role decomposition should not be treated as safety-neutral: it can act as a feature-conditioned architectural bias that changes evidence aggregation and shifts class-wise error profiles, indicating that prompt structure and architectural configuration should be explicitly tuned, validated, and reported in safety-sensitive LLM-based clinical decision systems.","abstract_license":"CC-BY (article license as recorded by Europe PMC; abstract reproduced unmodified apart from markup removal)","license":"CC-BY","is_open_access":false,"retracted":false,"peer_reviewed":false,"categories":[],"details":{"peer_reviewed":false,"note":"Preprint: not peer reviewed."},"provenance":"europepmc","attribution":"Abstract of \"Agent Role Structure and Operating Characteristics in Large Language Model Clinical Classification: A Comparative Study of Specialist and Deliberative Multi-Agent Protocols\" by Anderson C, https://doi.org/10.64898/2026.02.22.26346818, used under its Creative Commons Attribution license (CC-BY); markup removed, otherwise unmodified.","topics":[{"slug":"ai-medicine","name":"AI in Clinical Medicine","domain_slug":"medicine"}],"cited_in_dossiers":[]}