Reference decisions enhance LLM performance, amplified by source disclosure

Objective The rapid integration of large language models (LLMs) has propelled advancements in automated dialog technologies, improving the public's access to healthcare services. Drawing inspiration from the collaborative decision-making practices of medical professionals in complex cases, we i...

Full description

Saved in:
Bibliographic Details
Main Authors: Yongxiang Zhang, Zhaobin Liu, Shaosen Bai, Ting Xu, Raymond YK Lau
Format: Article
Language:English
Published: SAGE Publishing 2025-05-01
Series:Digital Health
Online Access:https://doi.org/10.1177/20552076251342078
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850269448547598336
author Yongxiang Zhang
Zhaobin Liu
Shaosen Bai
Ting Xu
Raymond YK Lau
author_facet Yongxiang Zhang
Zhaobin Liu
Shaosen Bai
Ting Xu
Raymond YK Lau
author_sort Yongxiang Zhang
collection DOAJ
description Objective The rapid integration of large language models (LLMs) has propelled advancements in automated dialog technologies, improving the public's access to healthcare services. Drawing inspiration from the collaborative decision-making practices of medical professionals in complex cases, we investigated whether LLMs could enhance their diagnostic accuracy through interaction. Methods An experimental study was conducted in China (September–December 2024) to investigate the impact of LLM-generated reference decisions and source disclosure on LLMs’ diagnostic performance. We used a Chinese clinical diagnostic task in a controlled comparative design, where three Chinese LLMs interpreted symptoms and conditions based on patient queries. LLMs’ outcomes were evaluated through accuracy and weighted F1 score metrics, with statistical analysis to determine significance. Results Analysis of variance on LLMs’ diagnostic accuracy scores demonstrated that incorporating LLM-generated decisions as a reference significantly improved diagnostic outcomes, with source disclosure amplifying this improvement. Conclusion Our findings underscore the potential of LLM collaboration in healthcare, offering strategies to refine response generation and decision-making across various applications.
format Article
id doaj-art-4eb4eb33a2cb4fb09d708c91413d5bb3
institution OA Journals
issn 2055-2076
language English
publishDate 2025-05-01
publisher SAGE Publishing
record_format Article
series Digital Health
spelling doaj-art-4eb4eb33a2cb4fb09d708c91413d5bb32025-08-20T01:53:08ZengSAGE PublishingDigital Health2055-20762025-05-011110.1177/20552076251342078Reference decisions enhance LLM performance, amplified by source disclosureYongxiang Zhang0Zhaobin Liu1Shaosen Bai2Ting Xu3Raymond YK Lau4 Department of Information Systems, , Hong Kong SAR, China Department of Information Systems, , Hong Kong SAR, China Department of Decisions, Operations and Technology, , Hong Kong SAR, China Department of Information Systems and Management Engineering, , Shenzhen, China Department of Information Systems, , Hong Kong SAR, ChinaObjective The rapid integration of large language models (LLMs) has propelled advancements in automated dialog technologies, improving the public's access to healthcare services. Drawing inspiration from the collaborative decision-making practices of medical professionals in complex cases, we investigated whether LLMs could enhance their diagnostic accuracy through interaction. Methods An experimental study was conducted in China (September–December 2024) to investigate the impact of LLM-generated reference decisions and source disclosure on LLMs’ diagnostic performance. We used a Chinese clinical diagnostic task in a controlled comparative design, where three Chinese LLMs interpreted symptoms and conditions based on patient queries. LLMs’ outcomes were evaluated through accuracy and weighted F1 score metrics, with statistical analysis to determine significance. Results Analysis of variance on LLMs’ diagnostic accuracy scores demonstrated that incorporating LLM-generated decisions as a reference significantly improved diagnostic outcomes, with source disclosure amplifying this improvement. Conclusion Our findings underscore the potential of LLM collaboration in healthcare, offering strategies to refine response generation and decision-making across various applications.https://doi.org/10.1177/20552076251342078
spellingShingle Yongxiang Zhang
Zhaobin Liu
Shaosen Bai
Ting Xu
Raymond YK Lau
Reference decisions enhance LLM performance, amplified by source disclosure
Digital Health
title Reference decisions enhance LLM performance, amplified by source disclosure
title_full Reference decisions enhance LLM performance, amplified by source disclosure
title_fullStr Reference decisions enhance LLM performance, amplified by source disclosure
title_full_unstemmed Reference decisions enhance LLM performance, amplified by source disclosure
title_short Reference decisions enhance LLM performance, amplified by source disclosure
title_sort reference decisions enhance llm performance amplified by source disclosure
url https://doi.org/10.1177/20552076251342078
work_keys_str_mv AT yongxiangzhang referencedecisionsenhancellmperformanceamplifiedbysourcedisclosure
AT zhaobinliu referencedecisionsenhancellmperformanceamplifiedbysourcedisclosure
AT shaosenbai referencedecisionsenhancellmperformanceamplifiedbysourcedisclosure
AT tingxu referencedecisionsenhancellmperformanceamplifiedbysourcedisclosure
AT raymondyklau referencedecisionsenhancellmperformanceamplifiedbysourcedisclosure