用多模态大模型自动生成可解释的分子语言，提升药物设计效率。
Foundation Molecular Grammar: Multi-Modal Foundation Models Induce Interpretable Molecular Graph Languages
- 利用多模态大模型将分子转为图像与文本，跨模态对齐生成可解释语言
- 在合成可行性、多样性与数据效率上均优于传统方法，无需人工标注
- 适合自动化分子发现流程，尤其适用于需要化学可解释性的研究者
近期的数据高效分子生成方法利用图语法提升生成模型的可解释性，但其语法学习依赖专家标注或不可靠启发式规则。本文提出基础分子语法（FMG），借助多模态基础模型（MMFMs）自动诱导可解释的分子语言。通过利用MMFM的化学知识，FMG将分子表示为图像，以文本描述，并通过提示学习对齐多模态信息。FMG可作为现有语法学习方法的直接替代方案，应用于分子生成与性质预测。实验表明，FMG不仅在合成可行性、多样性和数据效率方面表现优异，还内置化学可解释性，适用于自动化分子发现工作流。代码已开源：https://github.com/shiningsunnyday/induction。
原文摘要 · Abstract (English)
Recent data-efficient molecular generation approaches exploit graph grammars to introduce interpretability into the generative models. However, grammar learning therein relies on expert annotation or unreliable heuristics for algorithmic inference. We propose Foundation Molecular Grammar (FMG), which leverages multi-modal foundation models (MMFMs) to induce an interpretable molecular language. By exploiting the chemical knowledge of an MMFM, FMG renders molecules as images, describes them as text, and aligns information across modalities using prompt learning. FMG can be used as a drop-in replacement for the prior grammar learning approaches in molecular generation and property prediction. We show that FMG not only excels in synthesizability, diversity, and data efficiency but also offers built-in chemical interpretability for automated molecular discovery workflows. Code is available at https://github.com/shiningsunnyday/induction.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。