arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2403.01332q-bio.QMcs.AIcs.LG
收藏

链式思维与 LLM 链接以学习 DNA 结构生物物理学

Chaining thoughts and LLMs to learn DNA structural biophysics

MIT(麻省理工学院)

Tyler D. Ross, Ashwin Gopinath

AI总结:

本文通过微调 chatGPT 3.5-turbo,并采用链式思维推理与子任务模型链接方法,证明通用 LLM 可学习 DNA 结构生物物理规律并提升序列及结构分析设计能力。

摘要:

AI 科学家的未来发展,即一种能够整合多种实验数据并生成可检验假设的工具,具有巨大潜力。迄今为止,人们已经构建了定制的机器学习模型,专门用于单一科学任务,但除此之外缺乏通用模型的灵活性。在这里,我们展示了通用大型语言模型 chatGPT 3.5-turbo 可以经过微调来学习 DNA 的结构生物物理学。我们发现,无论是微调模型以返回链式思维响应,还是将针对子任务微调的模型链接在一起,都能增强分析和设计 DNA 序列及其结构的能力。

英文摘要:

The future development of an AI scientist, a tool that is capable of integrating a variety of experimental data and generating testable hypotheses, holds immense potential. So far, bespoke machine learning models have been created to specialize in singular scientific tasks, but otherwise lack the flexibility of a general purpose model. Here, we show that a general purpose large language model, chatGPT 3.5-turbo, can be fine-tuned to learn the structural biophysics of DNA. We find that both fine-tuning models to return chain-of-thought responses and chaining together models fine-tuned for subtasks have an enhanced ability to analyze and design DNA sequences and their structures.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑