评估大型语言模型的化学智能
Nicholas T Runcie1, Charlotte M Deane1, Fergus Imrie1
1Department of Statistics, University of Oxford, Oxford OX1 3LB, U.K.
Journal of chemical information and modeling
|December 18, 2025
概括
高级推理 大型语言模型 (LLM) 在化学任务中取得了显著进展,在新型ChemIQ基准上达到50-57%的准确性. 这些模型现在可以解释复杂的化学数据,反映人类化学家的推理.
科学领域:
- 人工智能的人工智能
- 计算化学计算化学
- 有机化学 有机化学
背景情况:
- 大型语言模型 (LLM) 在各种领域都表现出了多功能性.
- "推理模型"的出现在数学和软件工程等复杂问题解决领域显著增强了LLM的能力.
研究的目的:
- 在没有外部工具的情况下,评估推理模型对化学任务的直接应用.
- 引入一个新的基准,ChemIQ,用于评估核心有机化学概念和化学推理.
主要方法:
- 开发了ChemIQ基准标准,其中包括816个简短答案问题,重点是分子理解和化学推理.
- 在ChemIQ基准上对推理模型 (OpenAI的o3-mini,谷歌的Gemini 2.5 Pro,DeepSeek R1) 的评估.
- 评估模型性能,将SMILES字符串转换为IUPAC名称,并从NMR数据中阐明结构.
主要成果:
- 推理模型在ChemIQ基准上实现了50-57%的准确性,比非推理模型 (3-7%准确性) 显著改善.
- 模型展示了将SMILES字符串转换为IUPAC名称的能力,并从1D/2D1H和13C的NMR数据中阐明分子结构.
- 双子 2.5 Pro 精确生成了 ~ 90% 的 10 个重原子的分子的 SMILES 字符串,其中包括一个 25 个重原子的分子.
结论:
- 最新的推理模型显示,对高级化学推理任务的熟练程度越来越高.
- 这些模型使用的推理过程似乎模仿了人类化学家的推理过程.
- 在复杂的化学问题解决中,LLM正在成为直接应用的有能力的工具.
相关概念视频
Language and Cognition
688
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.
688
Higher Mental Functions of the Brain: Language
3.3K
Language is a system of communication that allows the expression of thoughts, ideas, and feelings. The brain processes language in both hemispheres.
Language formation and comprehension take place in the dominant hemisphere. The dominant hemisphere is responsible for understanding the meaning of spoken, written, or sign language, as well as the ability to communicate. For most people, the left hemisphere is the dominant one. The right hemisphere, then, gives tone and emotional context to the...
Language formation and comprehension take place in the dominant hemisphere. The dominant hemisphere is responsible for understanding the meaning of spoken, written, or sign language, as well as the ability to communicate. For most people, the left hemisphere is the dominant one. The right hemisphere, then, gives tone and emotional context to the...
3.3K
Molecular Models
43.4K
Physical models representing molecular architectures of chemical compounds play essential roles in understanding chemistry. The use of molecular models makes it easier to visualize the structures and shapes of atoms and molecules.
43.4K
Language Development
804
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...
804
Components of Language
721
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.
721
Chemical Synapses
11.1K
Chemical synapses are specialized sites between two neurons or between a neuron and a non-neuronal cell like a muscle, glandular or sensory cell.
Because chemical synapses depend on the release of neurotransmitter molecules from synaptic vesicles to pass on their signal, there is an approximately one millisecond delay between when the axon potential reaches the presynaptic terminal and when the neurotransmitter leads to opening of postsynaptic ion channels. Additionally, this signaling is...
Because chemical synapses depend on the release of neurotransmitter molecules from synaptic vesicles to pass on their signal, there is an approximately one millisecond delay between when the axon potential reaches the presynaptic terminal and when the neurotransmitter leads to opening of postsynaptic ion channels. Additionally, this signaling is...
11.1K

