KnowMol: Advancing Molecular Large Language Models with Multi-Level Chemical Knowledge

Zaifei Yang, Hong Chang, Ruibing Hou, Shiguang Shan, Xilin Chen · 2025

Read KnowMol: Advancing Molecular Large Language Models with Multi-Level Chemical Knowledge by Zaifei Yang, Hong Chang, Ruibing Hou, Shiguang Shan, Xilin Chen (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2510-19484.txt.

Description

Zaifei Yang and four co-authors present KnowMol, a large language model built to understand and generate molecules. Models of this kind have been held back by weak written descriptions of molecules and by clumsy ways of encoding molecular structure during training. The team attacks both problems at once.

They built KnowMol-100K, a dataset of one hundred thousand molecules with detailed annotations written at several levels of chemical description. They also designed a chemically informed way of representing molecules for the model. Trained on these, KnowMol posted strong results across molecular understanding and generation tasks, and the dataset is released publicly.

Catalog record

CREATORZaifei Yang, Hong Chang, Ruibing Hou, Shiguang Shan, Xilin Chen
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE78 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2510-19484

Browse the full library · About the Tower of Babel