MedCite: Can Language Models Generate Verifiable Text for Medicine?

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Wang, Xiao, Tan, Mengjue, Jin, Qiao, Xiong, Guangzhi, Hu, Yu, Zhang, Aidong, Lu, Zhiyong, Zhang, Minjia
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866908397722402816
author Wang, Xiao
Tan, Mengjue
Jin, Qiao
Xiong, Guangzhi
Hu, Yu
Zhang, Aidong
Lu, Zhiyong
Zhang, Minjia
author_facet Wang, Xiao
Tan, Mengjue
Jin, Qiao
Xiong, Guangzhi
Hu, Yu
Zhang, Aidong
Lu, Zhiyong
Zhang, Minjia
contents Existing LLM-based medical question-answering systems lack citation generation and evaluation capabilities, raising concerns about their adoption in practice. In this work, we introduce \name, the first end-to-end framework that facilitates the design and evaluation of citation generation with LLMs for medical tasks. Meanwhile, we introduce a novel multi-pass retrieval-citation method that generates high-quality citations. Our evaluation highlights the challenges and opportunities of citation generation for medical tasks, while identifying important design choices that have a significant impact on the final citation quality. Our proposed method achieves superior citation precision and recall improvements compared to strong baseline methods, and we show that evaluation results correlate well with annotation results from professional experts.
format Preprint
id arxiv_https___arxiv_org_abs_2506_06605
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MedCite: Can Language Models Generate Verifiable Text for Medicine?
Wang, Xiao
Tan, Mengjue
Jin, Qiao
Xiong, Guangzhi
Hu, Yu
Zhang, Aidong
Lu, Zhiyong
Zhang, Minjia
Computation and Language
Artificial Intelligence
Existing LLM-based medical question-answering systems lack citation generation and evaluation capabilities, raising concerns about their adoption in practice. In this work, we introduce \name, the first end-to-end framework that facilitates the design and evaluation of citation generation with LLMs for medical tasks. Meanwhile, we introduce a novel multi-pass retrieval-citation method that generates high-quality citations. Our evaluation highlights the challenges and opportunities of citation generation for medical tasks, while identifying important design choices that have a significant impact on the final citation quality. Our proposed method achieves superior citation precision and recall improvements compared to strong baseline methods, and we show that evaluation results correlate well with annotation results from professional experts.
title MedCite: Can Language Models Generate Verifiable Text for Medicine?
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2506.06605