Large Language Models are In-Context Molecule Learners

arXiv:2403.04197v1 Announce Type: new
Abstract: Large Language Models (LLMs) have demonstrated exceptional performance in biochemical tasks, especially the molecule caption translation task, which aims to bridge the gap between molecules and natural language texts. However, previous methods in adapting LLMs to the molecule-caption translation task required extra domain-specific pre-training stages, suffered weak alignment between molecular and textual spaces, or imposed stringent demands on the scale of LLMs. To resolve the challenges, we propose In-Context Molecule Adaptation (ICMA), as a new …

abstract alignment arxiv bridge context cs.ai cs.cl domain extra gap however language language models large language large language models llms molecules natural natural language performance pre-training tasks training translation type