Your browser doesn't support javascript.
loading
Large Language Models, scientific knowledge and factuality: A framework to streamline human expert evaluation.
Wysocka, Magdalena; Wysocki, Oskar; Delmas, Maxime; Mutel, Vincent; Freitas, André.
Afiliación
  • Wysocka M; Digital Cancer Research, CRUK National Biomarker Centre, Manchester, United Kingdom; Department of Computer Science, University of Manchester, Manchester, United Kingdom. Electronic address: magdalena.wysocka@cruk.manchester.ac.uk.
  • Wysocki O; Digital Cancer Research, CRUK National Biomarker Centre, Manchester, United Kingdom; Idiap Research Institute, Martigny, Switzerland.
  • Delmas M; Idiap Research Institute, Martigny, Switzerland.
  • Mutel V; Inflamalps SA, Monthey, Switzerland.
  • Freitas A; Digital Cancer Research, CRUK National Biomarker Centre, Manchester, United Kingdom; Department of Computer Science, University of Manchester, Manchester, United Kingdom; Idiap Research Institute, Martigny, Switzerland.
J Biomed Inform ; 158: 104724, 2024 Oct.
Article en En | MEDLINE | ID: mdl-39277154
ABSTRACT

OBJECTIVE:

The paper introduces a framework for the evaluation of the encoding of factual scientific knowledge, designed to streamline the manual evaluation process typically conducted by domain experts. Inferring over and extracting information from Large Language Models (LLMs) trained on a large corpus of scientific literature can potentially define a step change in biomedical discovery, reducing the barriers for accessing and integrating existing medical evidence. This work explores the potential of LLMs for dialoguing with biomedical background knowledge, using the context of antibiotic discovery.

METHODS:

The framework involves three evaluation steps, each assessing different aspects sequentially fluency, prompt alignment, semantic coherence, factual knowledge, and specificity of the generated responses. By splitting these tasks between non-experts and experts, the framework reduces the effort required from the latter. The work provides a systematic assessment on the ability of eleven state-of-the-art LLMs, including ChatGPT, GPT-4 and Llama 2, in two prompting-based tasks chemical compound definition generation and chemical compound-fungus relation determination.

RESULTS:

Although recent models have improved in fluency, factual accuracy is still low and models are biased towards over-represented entities. The ability of LLMs to serve as biomedical knowledge bases is questioned, and the need for additional systematic evaluation frameworks is highlighted.

CONCLUSION:

While LLMs are currently not fit for purpose to be used as biomedical factual knowledge bases in a zero-shot setting, there is a promising emerging property in the direction of factuality as the models become domain specialised, scale up in size and level of human feedback.
Asunto(s)
Palabras clave

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Asunto principal: Procesamiento de Lenguaje Natural Límite: Humans Idioma: En Revista: J Biomed Inform Asunto de la revista: INFORMATICA MEDICA Año: 2024 Tipo del documento: Article Pais de publicación: Estados Unidos

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Asunto principal: Procesamiento de Lenguaje Natural Límite: Humans Idioma: En Revista: J Biomed Inform Asunto de la revista: INFORMATICA MEDICA Año: 2024 Tipo del documento: Article Pais de publicación: Estados Unidos