Beyond Accuracy: Investigating Error Types in GPT-4 Responses to USMLE Questions
Publikation: Beiträge in Sammelwerken › Aufsätze in Konferenzbänden › Forschung › begutachtet
Standard
SIGIR 2024 - Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval. Hrsg. / Grace Hui Yang; Hongning Wang; Sam Han; Claudia Hauff; Guido Zuccon; Yi Zhang. Association for Computing Machinery, Inc, 2024. S. 1073-1082 (SIGIR 2024 - Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval).
Publikation: Beiträge in Sammelwerken › Aufsätze in Konferenzbänden › Forschung › begutachtet
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - CHAP
T1 - Beyond Accuracy
T2 - 47th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2024
AU - Roy, Soumyadeep
AU - Khatua, Aparup
AU - Ghoochani, Fatemeh
AU - Hadler, Uwe
AU - Nejdl, Wolfgang
AU - Ganguly, Niloy
N1 - Publisher Copyright: © 2024 Owner/Author.
PY - 2024/7/10
Y1 - 2024/7/10
N2 - GPT-4 demonstrates high accuracy in medical QA tasks, leading with an accuracy of 86.70%, followed by Med-PaLM 2 at 86.50%. However, around 14% of errors remain. Additionally, current works use GPT-4 to only predict the correct option without providing any explanation and thus do not provide any insight into the thinking process and reasoning used by GPT-4 or other LLMs. Therefore, we introduce a new domain-specific error taxonomy derived from collaboration with medical students. Our GPT-4 USMLE Error (G4UE) dataset comprises 4153 GPT-4 correct responses and 919 incorrect responses to the United States Medical Licensing Examination (USMLE) respectively. These responses are quite long (258 words on average), containing detailed explanations from GPT-4 justifying the selected option. We then launch a large-scale annotation study using the Potato annotation platform and recruit 44 medical experts through Prolific, a well-known crowdsourcing platform. We annotated 300 out of these 919 incorrect data points at a granular level for different classes and created a multi-label span to identify the reasons behind the error. In our annotated dataset, a substantial portion of GPT-4's incorrect responses is categorized as a "Reasonable response by GPT-4,"by annotators. This sheds light on the challenge of discerning explanations that may lead to incorrect options, even among trained medical professionals. We also provide medical concepts and medical semantic predications extracted using the SemRep tool for every data point. We believe that it will aid in evaluating the ability of LLMs to answer complex medical questions. We make the resources available at https://github.com/roysoumya/usmle-gpt4-error-taxonomy.
AB - GPT-4 demonstrates high accuracy in medical QA tasks, leading with an accuracy of 86.70%, followed by Med-PaLM 2 at 86.50%. However, around 14% of errors remain. Additionally, current works use GPT-4 to only predict the correct option without providing any explanation and thus do not provide any insight into the thinking process and reasoning used by GPT-4 or other LLMs. Therefore, we introduce a new domain-specific error taxonomy derived from collaboration with medical students. Our GPT-4 USMLE Error (G4UE) dataset comprises 4153 GPT-4 correct responses and 919 incorrect responses to the United States Medical Licensing Examination (USMLE) respectively. These responses are quite long (258 words on average), containing detailed explanations from GPT-4 justifying the selected option. We then launch a large-scale annotation study using the Potato annotation platform and recruit 44 medical experts through Prolific, a well-known crowdsourcing platform. We annotated 300 out of these 919 incorrect data points at a granular level for different classes and created a multi-label span to identify the reasons behind the error. In our annotated dataset, a substantial portion of GPT-4's incorrect responses is categorized as a "Reasonable response by GPT-4,"by annotators. This sheds light on the challenge of discerning explanations that may lead to incorrect options, even among trained medical professionals. We also provide medical concepts and medical semantic predications extracted using the SemRep tool for every data point. We believe that it will aid in evaluating the ability of LLMs to answer complex medical questions. We make the resources available at https://github.com/roysoumya/usmle-gpt4-error-taxonomy.
KW - gpt-4
KW - medical qa
KW - multi-label dataset
KW - usmle error taxonomy
KW - Informatics
KW - Business informatics
UR - http://www.scopus.com/inward/record.url?scp=85199188807&partnerID=8YFLogxK
U2 - 10.1145/3626772.3657882
DO - 10.1145/3626772.3657882
M3 - Article in conference proceedings
AN - SCOPUS:85199188807
T3 - SIGIR 2024 - Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval
SP - 1073
EP - 1082
BT - SIGIR 2024 - Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval
A2 - Hui Yang, Grace
A2 - Wang, Hongning
A2 - Han, Sam
A2 - Hauff, Claudia
A2 - Zuccon, Guido
A2 - Zhang, Yi
PB - Association for Computing Machinery, Inc
Y2 - 14 July 2024 through 18 July 2024
ER -