A naukri.com initiative
Arxiv
6d
8
Image Credit: Arxiv
Large Language Models for Controllable Multi-property Multi-objective Molecule Optimization
- Researchers have introduced C-MuMOInstruct, a dataset focused on multi-property optimization in drug design.
- GeLLMO-Cs, a series of instruction-tuned Large Language Models (LLMs), have been developed to perform targeted property-specific optimization.
- Experiments across tasks show that GeLLMO-Cs outperform strong baselines, with up to 126% higher success rate.
- GeLLMO-Cs demonstrate impressive generalization to novel tasks and unseen instructions, aiming to support diverse optimizations in drug design.
Read Full Article
Like
For uninterrupted reading, download the app