Advancing Scientific Discovery with Large Language Models and Knowledge Graphs
We propose ERAST, an LLM-empowered homology search tool with vector databases for ultra-fast, high-precision biological sequence retrieval at billion-scale.
This survey summarizes SciKG construction, cross-domain applications and LLM collaboration, and prospects advanced SciKG ecosystems for scientific discovery.
We present InstructBioMol, a large language model designed to bridge natural language and biomolecules through any-to-any alignments.
We present SciToolAgent, an LLM-powered agent that automates hundreds of scientific tools across biology, chemistry, and materials science.
We propose a method for knowledge graph-enhanced molecular contrastive learning with functional prompt (KANO).
Biology
Chemistry
Physics
Materials
Medicine