ScienceMind
Advancing Scientific Discovery with Large Language Models and Knowledge Graphs
Evaluating OpenAI o1 with SciKnowEval
We release an evaluation report about the capabilities of o1 in scientific domains.
Retrosynthesis Prediction with An Iterative String Editing Model
We propose reframing single-step retrosynthesis prediction as a molecular string editing task, iteratively refining target molecule strings to generate precursor compounds.
Scientific Large Language Models: A Survey on Biological & Chemical Domains
We release a survey paper about scientific large language models, focusing on textual, molecular, protein and genomic languages, as well as their combination.
Knowledge Graph-enhanced Molecular Contrastive Learning with Functional Prompt
We propose a method for knowledge graph-enhanced molecular contrastive learning with functional prompt (KANO).
Learning Invariant Molecular Representation in Latent Discrete Space
We propose a new molecular representations learning framework that exhibit invariance and robustness against distribution shifts.
Biology
Chemistry
Physics
Materials
Medicine