_version_ 1866910375086129152
author Mukherjee, Subhabrata
Gamble, Paul
Ausin, Markel Sanz
Kant, Neel
Aggarwal, Kriti
Manjunath, Neha
Datta, Debajyoti
Liu, Zhengliang
Ding, Jiayuan
Busacca, Sophia
Bianco, Cezanne
Sharma, Swapnil
Lasko, Rae
Voisard, Michelle
Harneja, Sanchay
Filippova, Darya
Meixiong, Gerry
Cha, Kevin
Youssefi, Amir
Buvanesh, Meyhaa
Weingram, Howard
Bierman-Lytle, Sebastian
Mangat, Harpreet Singh
Parikh, Kim
Godil, Saad
Miller, Alex
author_facet Mukherjee, Subhabrata
Gamble, Paul
Ausin, Markel Sanz
Kant, Neel
Aggarwal, Kriti
Manjunath, Neha
Datta, Debajyoti
Liu, Zhengliang
Ding, Jiayuan
Busacca, Sophia
Bianco, Cezanne
Sharma, Swapnil
Lasko, Rae
Voisard, Michelle
Harneja, Sanchay
Filippova, Darya
Meixiong, Gerry
Cha, Kevin
Youssefi, Amir
Buvanesh, Meyhaa
Weingram, Howard
Bierman-Lytle, Sebastian
Mangat, Harpreet Singh
Parikh, Kim
Godil, Saad
Miller, Alex
contents We develop Polaris, the first safety-focused LLM constellation for real-time patient-AI healthcare conversations. Unlike prior LLM works in healthcare focusing on tasks like question answering, our work specifically focuses on long multi-turn voice conversations. Our one-trillion parameter constellation system is composed of several multibillion parameter LLMs as co-operative agents: a stateful primary agent that focuses on driving an engaging conversation and several specialist support agents focused on healthcare tasks performed by nurses to increase safety and reduce hallucinations. We develop a sophisticated training protocol for iterative co-training of the agents that optimize for diverse objectives. We train our models on proprietary data, clinical care plans, healthcare regulatory documents, medical manuals, and other medical reasoning documents. We align our models to speak like medical professionals, using organic healthcare conversations and simulated ones between patient actors and experienced nurses. This allows our system to express unique capabilities such as rapport building, trust building, empathy and bedside manner. Finally, we present the first comprehensive clinician evaluation of an LLM system for healthcare. We recruited over 1100 U.S. licensed nurses and over 130 U.S. licensed physicians to perform end-to-end conversational evaluations of our system by posing as patients and rating the system on several measures. We demonstrate Polaris performs on par with human nurses on aggregate across dimensions such as medical safety, clinical readiness, conversational quality, and bedside manner. Additionally, we conduct a challenging task-based evaluation of the individual specialist support agents, where we demonstrate our LLM agents significantly outperform a much larger general-purpose LLM (GPT-4) as well as from its own medium-size class (LLaMA-2 70B).
format Preprint
id arxiv_https___arxiv_org_abs_2403_13313
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Polaris: A Safety-focused LLM Constellation Architecture for Healthcare
Mukherjee, Subhabrata
Gamble, Paul
Ausin, Markel Sanz
Kant, Neel
Aggarwal, Kriti
Manjunath, Neha
Datta, Debajyoti
Liu, Zhengliang
Ding, Jiayuan
Busacca, Sophia
Bianco, Cezanne
Sharma, Swapnil
Lasko, Rae
Voisard, Michelle
Harneja, Sanchay
Filippova, Darya
Meixiong, Gerry
Cha, Kevin
Youssefi, Amir
Buvanesh, Meyhaa
Weingram, Howard
Bierman-Lytle, Sebastian
Mangat, Harpreet Singh
Parikh, Kim
Godil, Saad
Miller, Alex
Artificial Intelligence
Computation and Language
We develop Polaris, the first safety-focused LLM constellation for real-time patient-AI healthcare conversations. Unlike prior LLM works in healthcare focusing on tasks like question answering, our work specifically focuses on long multi-turn voice conversations. Our one-trillion parameter constellation system is composed of several multibillion parameter LLMs as co-operative agents: a stateful primary agent that focuses on driving an engaging conversation and several specialist support agents focused on healthcare tasks performed by nurses to increase safety and reduce hallucinations. We develop a sophisticated training protocol for iterative co-training of the agents that optimize for diverse objectives. We train our models on proprietary data, clinical care plans, healthcare regulatory documents, medical manuals, and other medical reasoning documents. We align our models to speak like medical professionals, using organic healthcare conversations and simulated ones between patient actors and experienced nurses. This allows our system to express unique capabilities such as rapport building, trust building, empathy and bedside manner. Finally, we present the first comprehensive clinician evaluation of an LLM system for healthcare. We recruited over 1100 U.S. licensed nurses and over 130 U.S. licensed physicians to perform end-to-end conversational evaluations of our system by posing as patients and rating the system on several measures. We demonstrate Polaris performs on par with human nurses on aggregate across dimensions such as medical safety, clinical readiness, conversational quality, and bedside manner. Additionally, we conduct a challenging task-based evaluation of the individual specialist support agents, where we demonstrate our LLM agents significantly outperform a much larger general-purpose LLM (GPT-4) as well as from its own medium-size class (LLaMA-2 70B).
title Polaris: A Safety-focused LLM Constellation Architecture for Healthcare
topic Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2403.13313