Biomedical Large Languages Models Seem not to be Superior to Generalist Models on Unseen Medical Data

• URL: http://arxiv.org/abs/2408.13833v1
• Date: Sun, 25 Aug 2024 13:36:22 GMT
• Title: Biomedical Large Languages Models Seem not to be Superior to Generalist Models on Unseen Medical Data
• Authors: Felix J. Dorfner, Amin Dada, Felix Busch, Marcus R. Makowski, Tianyu Han, Daniel Truhn, Jens Kleesiek, Madhumita Sushil, Jacqueline Lammert, Lisa C. Adams, Keno K. Bressem,
• Abstract summary: Large language models (LLMs) have shown potential in biomedical applications, leading to efforts to fine-tune them on domain-specific data. This study evaluates the performance of biomedically fine-tuned LLMs against their general-purpose counterparts on a variety of clinical tasks.
• Score: 3.469567586411153
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models (LLMs) have shown potential in biomedical applications, leading to efforts to fine-tune them on domain-specific data. However, the effectiveness of this approach remains unclear. This study evaluates the performance of biomedically fine-tuned LLMs against their general-purpose counterparts on a variety of clinical tasks. We evaluated their performance on clinical case challenges from the New England Journal of Medicine (NEJM) and the Journal of the American Medical Association (JAMA) and on several clinical tasks (e.g., information extraction, document summarization, and clinical coding). Using benchmarks specifically chosen to be likely outside the fine-tuning datasets of biomedical models, we found that biomedical LLMs mostly perform inferior to their general-purpose counterparts, especially on tasks not focused on medical knowledge. While larger models showed similar performance on case tasks (e.g., OpenBioLLM-70B: 66.4% vs. Llama-3-70B-Instruct: 65% on JAMA cases), smaller biomedical models showed more pronounced underperformance (e.g., OpenBioLLM-8B: 30% vs. Llama-3-8B-Instruct: 64.3% on NEJM cases). Similar trends were observed across the CLUE (Clinical Language Understanding Evaluation) benchmark tasks, with general-purpose models often performing better on text generation, question answering, and coding tasks. Our results suggest that fine-tuning LLMs to biomedical data may not provide the expected benefits and may potentially lead to reduced performance, challenging prevailing assumptions about domain-specific adaptation of LLMs and highlighting the need for more rigorous evaluation frameworks in healthcare AI. Alternative approaches, such as retrieval-augmented generation, may be more effective in enhancing the biomedical capabilities of LLMs without compromising their general knowledge.

Related papers

• GMAI-MMBench: A Comprehensive Multimodal Evaluation Benchmark Towards General Medical AI [67.09501109871351]
  Large Vision-Language Models (LVLMs) are capable of handling diverse data types such as imaging, text, and physiological signals. GMAI-MMBench is the most comprehensive general medical AI benchmark with well-categorized data structure and multi-perceptual granularity to date. It is constructed from 284 datasets across 38 medical image modalities, 18 clinical-related tasks, 18 departments, and 4 perceptual granularities in a Visual Question Answering (VQA) format.
  arXiv  Detail & Related papers  (2024-08-06T17:59:21Z)
• SemioLLM: Assessing Large Language Models for Semiological Analysis in Epilepsy Research [45.2233252981348]
  Large Language Models have shown promising results in their ability to encode general medical knowledge. We test the ability of state-of-the-art LLMs to leverage their internal knowledge and reasoning for epilepsy diagnosis.
  arXiv  Detail & Related papers  (2024-07-03T11:02:12Z)
• BiomedRAG: A Retrieval Augmented Large Language Model for Biomedicine [19.861178160437827]
  Large Language Models (LLMs) have swiftly emerged as vital resources for different applications in the biomedical and healthcare domains. textscBiomedRAG attains superior performance across 5 biomedical NLP tasks. textscBiomedRAG outperforms other triple extraction systems with micro-F1 scores of 81.42 and 88.83 on GIT and ChemProt corpora, respectively.
  arXiv  Detail & Related papers  (2024-05-01T12:01:39Z)
• Does Biomedical Training Lead to Better Medical Performance? [2.3814275542331385]
  Large Language Models (LLMs) are expected to significantly contribute to patient care, diagnostics, and administrative processes. This study investigates the effect of biomedical training in the context of six practical medical tasks evaluating $25$ models.
  arXiv  Detail & Related papers  (2024-04-05T12:51:37Z)
• BioMedLM: A 2.7B Parameter Language Model Trained On Biomedical Text [82.7001841679981]
  BioMedLM is a 2.7 billion parameter GPT-style autoregressive model trained exclusively on PubMed abstracts and full articles. When fine-tuned, BioMedLM can produce strong multiple-choice biomedical question-answering results competitive with larger models. BioMedLM can also be fine-tuned to produce useful answers to patient questions on medical topics.
  arXiv  Detail & Related papers  (2024-03-27T10:18:21Z)
• Towards a clinically accessible radiology foundation model: open-access and lightweight, with automated evaluation [113.5002649181103]
  Training open-source small multimodal models (SMMs) to bridge competency gaps for unmet clinical needs in radiology. For training, we assemble a large dataset of over 697 thousand radiology image-text pairs. For evaluation, we propose CheXprompt, a GPT-4-based metric for factuality evaluation, and demonstrate its parity with expert evaluation. The inference of LlaVA-Rad is fast and can be performed on a single V100 GPU in private settings, offering a promising state-of-the-art tool for real-world clinical applications.
  arXiv  Detail & Related papers  (2024-03-12T18:12:02Z)
• An Evaluation of Large Language Models in Bioinformatics Research [52.100233156012756]
  We study the performance of large language models (LLMs) on a wide spectrum of crucial bioinformatics tasks. These tasks include the identification of potential coding regions, extraction of named entities for genes and proteins, detection of antimicrobial and anti-cancer peptides, molecular optimization, and resolution of educational bioinformatics problems. Our findings indicate that, given appropriate prompts, LLMs like GPT variants can successfully handle most of these tasks.
  arXiv  Detail & Related papers  (2024-02-21T11:27:31Z)
• A Comprehensive Evaluation of Large Language Models on Benchmark Biomedical Text Processing Tasks [2.5027382653219155]
  This paper aims to evaluate the performance of Large Language Models (LLM) on benchmark biomedical tasks. To the best of our knowledge, this is the first work that conducts an extensive evaluation and comparison of various LLMs in the biomedical domain.
  arXiv  Detail & Related papers  (2023-10-06T14:16:28Z)
• BiomedGPT: A Generalist Vision-Language Foundation Model for Diverse Biomedical Tasks [68.39821375903591]
  Generalist AI holds the potential to address limitations due to its versatility in interpreting different data types. Here, we propose BiomedGPT, the first open-source and lightweight vision-language foundation model.
  arXiv  Detail & Related papers  (2023-05-26T17:14:43Z)
• Evaluation of ChatGPT Family of Models for Biomedical Reasoning and Classification [6.163540203358258]
  This study investigates the performance of large language models (LLMs) in biomedical tasks beyond question-answering. Because no patient data can be passed to the OpenAI API public interface, we evaluated model performance with over 10000 samples. We found that fine-tuning for two fundamental NLP tasks remained the best strategy.
  arXiv  Detail & Related papers  (2023-04-05T15:11:25Z)
• Federated Learning Enables Big Data for Rare Cancer Boundary Detection [98.5549882883963]
  We present findings from the largest Federated ML study to-date, involving data from 71 healthcare institutions across 6 continents. We generate an automatic tumor boundary detector for the rare disease of glioblastoma. We demonstrate a 33% improvement over a publicly trained model to delineate the surgically targetable tumor, and 23% improvement over the tumor's entire extent.
  arXiv  Detail & Related papers  (2022-04-22T17:27:00Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.