模型调整还是提示调整? 临床概念和关系提取的大型语言模型的研究
Cheng Peng1, Xi Yang2, Kaleb E Smith3
1Department of Health Outcomes and Biomedical Informatics, College of Medicine, University of Florida, Gainesville, FL, USA.
Journal of biomedical informatics
|March 28, 2024
概括
基于软提示的学习可以提高临床应用中的大型语言模型 (LLM) 的性能. 具有软提示的冷LLM提供了竞争性的短暂学习和通用性,同时降低了计算成本.
科学领域:
- 人工智能的人工智能
- 自然语言处理自然语言处理.
- 计算语言学 计算语言学
背景情况:
- 大型语言模型 (LLM) 在各种领域显示出希望,但它们在临床文本分析等专业领域的应用需要有效的适应策略.
- 提示调整,一种在不需要广泛的再培训的情况下适应LLM的方法,是一个活跃的研究领域,软提示为人为工程硬提示提供了数据驱动的替代方案.
研究的目的:
- 为大型语言模型 (LLM) 开发和评估基于提示的软学习架构.
- 为了比较软提示与传统微调和硬提示方法的有效性,使用冷和未冷的LLMs.
- 在临床自然语言处理任务中评估拟议架构的转移学习和少量学习能力.
主要方法:
- 开发了一个基于软提示的学习架构,并与四种策略进行了比较:没有提示,用未结的LLMs进行硬提示,用未结的LLMs进行软提示,以及用结的LLMs进行软提示.
- 采用两个基准数据集 (不良药物事件和健康的社会决定因素) 来评估GatorTron临床LLM (高达89亿个参数) 的临床概念和关系提取任务.
- 评估了近距离学习能力和跨机构通用性.
主要成果:
- 使用未冷的GatorTron-3.9B软提示实现了概念提取 (0.9118) 和关系提取 (0.8332) 的优异F1得分,超过了传统的微调和硬提示.
- 虽然小型冷LLM表现不佳,但具有软提示的大型冷LLM (例如,GatorTron-8.9B) 显示出具有竞争力的表现,特别是在跨机构评估中.
- 与未结模型相比,具有软提示的冷LLM显著降低了计算成本 (2.56%),突出了其效率.
结论:
- 机器学习的软提示比人类设计的硬提示更有效,用于LLM适应.
- 结的LLM,特别是大规模的LLM,表现出强烈的少量学习和跨机构应用的概括性.
- 使用冷的LLM软提示式学习为临床NLP提供了一种高效和有效的方法,降低了计算需求,同时保持了高性能.
相关概念视频
Improving Translational Accuracy
10.3K
Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...
10.3K
Typical Model Studies
358
Fluid mechanics model studies often utilize scaled-down systems to predict fluid behavior in full-scale environments, such as river flows, dam spillways, and structures interacting with open surfaces. Maintaining Froude number similarity in river models is crucial, as it replicates surface flow features like wave patterns and velocities.
358
Concepts and Prototypes
139
The human nervous system handles vast amounts of information by translating sensory stimuli into neural impulses, which the brain processes, creating thoughts expressed through language or stored as memories. The brain also synthesizes information from emotions and memories, which significantly influence thoughts and behaviors. This intricate process creates a comprehensive mental picture.
The brain organizes this information using concepts, which are mental categories grouping linguistic data,...
The brain organizes this information using concepts, which are mental categories grouping linguistic data,...
139
Pharmacokinetic Models: Comparison and Selection Criterion
70
Physiological and compartmental models are valuable tools used in studying biological systems. These models rely on differential equations to maintain mass balance within the system, ensuring an accurate representation of the dynamic processes at play.
Physiological models take a detailed approach by considering specific molecular processes. They can predict drug distribution, metabolism, and elimination changes, providing a comprehensive understanding of how drugs interact with the body.
Physiological models take a detailed approach by considering specific molecular processes. They can predict drug distribution, metabolism, and elimination changes, providing a comprehensive understanding of how drugs interact with the body.
70
Modeling in Therapy
72
Modeling, a key technique in therapy, uses observational learning to help clients acquire and practice new skills by watching therapists demonstrate desired behaviors. This approach, rooted in Albert Bandura's concept of vicarious learning, plays a significant role in therapeutic interventions for various psychological conditions, including social anxiety, ADHD, and depression.
Participant Modeling
Participant modeling involves therapists demonstrating calm and effective behaviors in...
Participant Modeling
Participant modeling involves therapists demonstrating calm and effective behaviors in...
72
Generalization, Discrimination, and Extinction
548
Generalization, discrimination, and extinction are key concepts in operant conditioning that influence how behaviors are learned and maintained.
Generalization occurs when a behavior reinforced in one context is performed in similar situations. For instance, a student who studies diligently for calculus and receives excellent grades might apply the same study habits to psychology and history, expecting similar results. Generalization shows how learning in one setting can influence behavior in...
Generalization occurs when a behavior reinforced in one context is performed in similar situations. For instance, a student who studies diligently for calculus and receives excellent grades might apply the same study habits to psychology and history, expecting similar results. Generalization shows how learning in one setting can influence behavior in...
548


