评估传统中医药考试题的大规模和轻量级大语言模型:一项比较研究
Yizhen Li1, Shaohan Huang2, Jiaxing Qi2
1The First Affiliated Hospital of Henan University of Chinese Medicine, Zhengzhou, China.
Journal of evidence-based medicine
|February 11, 2026
概括
大型语言模型 (LLM) 在中国传统医学 (TCM) 中表现有前途,但在多选题问题上却很困难. 优化的轻量级LLM,如Qwen3-1.7B,为资源有限的医疗保健机构提供了可行的替代方案.
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.
- 传统中国医药 传统中国医药
背景情况:
- 传统中医 (TCM) 对大型语言模型 (LLM) 提出了独特的挑战,因为其知识密集型的性质.
- 虽然存在TCM特定的基准,但轻量级LLM在这个领域的表现尚未得到充分探索.
- 本研究系统地评估和比较TCM应用的大规模和轻量级LLM,考虑部署的权衡.
研究的目的:
- 系统地评估和比较各种大型和轻量级LLM在TCM相关的问答任务上的表现.
- 在TCM的背景下,评估不同LLM大小的能力和部署权衡.
- 确定影响LLM绩效的因素,如提示策略和语言.
主要方法:
- 从教科书中开发了一个801个TCM相关问题的数据集.
- 用英语和中文的零射击和少数射击提示来评估11个LLM.
- 绩效主要以准确度来衡量,重点关注不同的问题类型和推理类别.
主要成果:
- 大规模的LLM在单选题和真假题中表现出色,但在多选题中表现不佳.
- 轻量级LLM通常落后于较大的型号,Qwen3-1.7B表现出具有竞争力的性能,甚至超过了专门的TCMChat-7B.
- 短暂的提示和中文提示通常改善了LLM的表现,而症状诊断被证明是最具挑战性的推理任务.
结论:
- 大规模的LLM在TCM中显示出强大的知识回忆力,但它们在多选项任务中的局限性和高计算成本可能会阻碍临床应用.
- 优化的轻量级LLM,以Qwen3-1.7B为例,表明模型优化和特定领域的培训比简单地增加模型大小更有利.
- 结果为在资源有限的医疗环境中部署优化的LLM提供了洞察力,尽管在现实世界的临床决策中需要进一步评估.
相关概念视频
01:16Language
921
Language is a unique communication system that uses words and systematic rules to organize and transmit information. Unlike other forms of communication, which may involve postures, movements, odors, or vocalizations, language relies on symbols and grammar. This makes human communication distinct from that of other species, who also communicate but do not use language in the same way humans do.
Corballis and Suddendorf (2007) and Tomasello and Rakoczy (2003) highlight the role of language in...
Corballis and Suddendorf (2007) and Tomasello and Rakoczy (2003) highlight the role of language in...
921
02:41pH Scale
80.4K
Hydronium and hydroxide ions are present both in pure water and in all aqueous solutions, and their concentrations are inversely proportional as determined by the ion product of water (Kw). The concentrations of these ions in a solution are often critical determinants of the solution’s properties and the chemical behaviors of its other solutes. Two different solutions can differ in their hydronium or hydroxide ion concentrations by a million, billion, or even trillion times. A common means of...
80.4K
01:24Components of Language
830
Language, whether spoken, signed, or written, consists of specific components: lexicon and grammar. The lexicon is the vocabulary of a language, comprising its words. Grammar is the set of rules used to convey meaning through the lexicon. For example, English grammar adds “-ed” to most verbs to indicate past tense. Words are formed by combining phonemes, which are the basic sound units of a language. Different languages have different sets of phonemes (e.g., “ah” vs.
830
01:22Language Development
936
Children master language quickly and with relative ease, supported by both biological predisposition and reinforcement. B. F. Skinner (1957) proposed that language is learned through reinforcement, while Noam Chomsky (1965) argued that language acquisition mechanisms are biologically determined.
The critical period for language acquisition suggests that the ability to acquire language is at its peak early in life. As people age, this proficiency decreases. Language development begins very...
The critical period for language acquisition suggests that the ability to acquire language is at its peak early in life. As people age, this proficiency decreases. Language development begins very...
936
01:27Language and Cognition
812
Language serves as a bridge between ideas and communication, influencing how individuals perceive and interact with the world. Psychologists have long debated whether language shapes thought or vice versa. This discussion gained grip with Edward Sapir and Benjamin Lee Whorf in the 1940s, who proposed that language determines thought, a concept known as linguistic determinism. They suggested that the vocabulary and structure of a language influence how its speakers think and perceive reality.
812
01:29Self-Evaluation Maintenance Model
337
The Self-Evaluation Maintenance (SEM) model offers a psychological framework to understand how individuals’ self-esteem is influenced by the achievements of others, particularly those with whom they share close personal bonds. The SEM model operates when personal rather than social identity guides individuals. Central to this model is the notion that individuals have an inherent desire to preserve a favorable self-image, which is continuously shaped by interpersonal comparisons and...
337