arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2406.06777cs.CVcs.AI
收藏

MolX: 通过多模态扩展增强大型语言模型的分子理解能力

MolX: Enhancing Large Language Models for Molecular Understanding With A Multi-Modal Extension

University of Notre Dame, IN, USA(诺丁汉大学);University of California, Los Angeles, CA, USA(加州大学洛杉矶分校)

Khiem Le, Zhichun Guo, Kaiwen Dong, Xiaobao Huang, Bozhao Nan, Roshni Iyer, Xiangliang Zhang, Olaf Wiest, Wei Wang, Ting Hua, Nitesh V. Chawla

AI总结:

MolX通过多模态扩展提升LLM对分子的理解能力,利用SMILES和分子图提取细粒度特征,并结合分子指纹提升性能,有效提升分子相关任务表现。

摘要:

大型语言模型(LLMs)凭借其强大的任务处理能力,在多个领域展现出显著进展,已超越自然语言理解。然而,它们在化学领域的表现仍然受限,尤其是在解决分子相关任务时。这一挑战归因于其在仅使用常见文本表示(即SMILES字符串)来理解分子时的固有局限性。在本研究中,我们试图通过给LLMs添加一个多模态外部模块MolX来增强其理解分子的能力。我们不直接使用SMILES字符串来表示分子,而是利用特定编码器从SMILES字符串和2D分子图表示中提取细粒度特征,以供输入LLM。我们还引入了手工制作的分子指纹,以利用其嵌入的领域知识。为了建立MolX与LLM文本输入空间之间的对齐,冻结LLM的模型通过包含多样化任务的策略进行预训练。实验评估显示,我们的方法在从分子到文本翻译到分子属性预测等下游分子相关任务中,无论是否微调LLM,均优于基线方法,同时仅引入少量可训练参数-0.53%和0.82%。

英文摘要:

Large Language Models (LLMs) with their strong task-handling capabilities have shown remarkable advancements across a spectrum of fields, moving beyond natural language understanding. However, their proficiency within the chemistry domain remains restricted, especially in solving molecule-related tasks. This challenge is attributed to their inherent limitations in comprehending molecules using only common textual representations, i.e. SMILES strings. In this study, we seek to enhance the ability of LLMs to comprehend molecules by equipping them with a multi-modal external module, termed MolX. Instead of directly using SMILES strings to represent a molecule, we utilize specific encoders to extract fine-grained features from both SMILES string and 2D molecular graph representations for feeding into an LLM. A hand-crafted molecular fingerprint is incorporated to leverage its embedded domain knowledge. To establish an alignment between MolX and the LLM's textual input space, the model in which the LLM is frozen, is pre-trained with a strategy including a diverse set of tasks. Experimental evaluations show that our proposed method outperforms baselines across downstream molecule-related tasks ranging from molecule-to-text translation to molecular property prediction, with and without fine-tuning the LLM, while only introducing a small number of trainable parameters-0.53% and 0.82%, respectively.

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。

↑