This project focuses on applying large language models to solve chemistry-related problems.

• Conducted research on existing large language models in chemistry and molecule and related downstream tasks. Researched advancements in AI for molecular representation and retrosynthesis.

• The main objectives include:

  • Molecule name prediction
  • Molecule property prediction
  • Chemical yield prediction
  • Chemical reaction prediction
  • Chemical reagents selection
  • Chemical retrosynthesis
  • Text-based molecule design
  • Molecule captioning

• Developed scripts to process collected data according to the team’s needs and communicated with the team. Reimplemented the ChemLLMBench code to facilitate the team’s evaluation of our model.

Reference

• [1] What can Large Language Models do in chemistry? A comprehensive benchmark on eight tasks

This research project was conducted during my visiting at CUHK(Shenzhen) under the supervision of Prof. Benyou Wang.