Augmenting large language models with chemistry tools
Расширение возможностей больших языковых моделей с помощью химических инструментов
2024-05-08
SCID: 54.1/4t3bz2w9
Discuss with AI
ChemCrowchemistry agentdrug discoverylarge language modelsorganic synthesis
Figures from the paper
Abstract (AI)
Large language models (LLMs) have shown strong performance in tasks across domains but struggle with chemistry-related problems. These models also lack access to external knowledge sources, limiting their usefulness in scientific applications. We introduce ChemCrow, an LLM chemistry agent designed to accomplish tasks across organic synthesis, drug discovery and materials design. By integrating 18 expert-designed tools and using GPT-4 as the LLM, ChemCrow augments the LLM performance in chemistry, and new capabilities emerge. Our agent autonomously planned and executed the syntheses of an insect repellent and three organocatalysts and guided the discovery of a novel chromophore. Our evaluation, including both LLM and expert assessments, demonstrates ChemCrow's effectiveness in automating a diverse set of chemical tasks. Our work not only aids expert chemists and lowers barriers for non-experts but also fosters scientific advancement by bridging the gap between experimental and computational chemistry.
Key Findings
1
ChemCrow autonomously planned and executed syntheses of an insect repellent and three organocatalysts.
2
ChemCrow guided the discovery of a novel chromophore, demonstrating utility in materials-related research.
3
ChemCrow is a chemistry-focused LLM agent that integrates GPT-4 with 18 expert-designed tools for organic synthesis, drug discovery, and materials design.
4
LLM-based and expert evaluations indicate that ChemCrow can automate a diverse range of chemical tasks and support both experts and non-experts.
5
The tool-augmented agent develops chemistry capabilities beyond those available to the underlying language model alone.
Research Object
ChemCrow, an LLM chemistry agent augmented with 18 expert-designed tools
Research Subject
Automated execution and effectiveness across organic synthesis, drug discovery, materials design, and chemical task-solving, including planning syntheses and discovering a novel chromophore
Publication Details
Publication Date
2024-05-08
Journal
Publisher
ISSN
Cited by
792
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai8
LLaMA: Open and Efficient Foundation Language Models2023
Automatic Chemical Design Using a Data-Driven Continuous Representation of Molecules2018
Training language models to follow instructions with human feedback2022
MoleculeNet: a benchmark for molecular machine learning2017
On the Opportunities and Risks of Foundation Models2021
PaLM: Scaling Language Modeling with Pathways2022
Sparks of Artificial General Intelligence: Early experiments with GPT-42023
Autonomous chemical research with large language models2023
Cited by20
A survey on large language model based autonomous agents2024
Autonomous chemical research with large language models2023
Self-Driving Laboratories for Chemistry and Materials Science2024
A survey on LLM-based multi-agent systems: workflow, infrastructure, and challenges2024
Empowering biomedical discovery with AI agents2024
Visibility into AI Agents2024
Exploring Large Language Model‐Based Intelligent Agents: Definitions, Methods, and Prospects2026
A review of large language models and autonomous agents in chemistry2024
Autonomous ‘self-driving’ laboratories: a review of technology and policy implications2025
From language to action: a review of large language models as autonomous agents and tool users2026
Retrieval-Augmented Generation (RAG)2025
Applied Artificial Intelligence in Materials Science and Material Design2025
AI, agentic models and lab automation for scientific discovery — the beginning of scAInce2025
Risks of AI scientists: prioritizing safeguarding over autonomy2025
BioMaster: Multi-agent system for automated bioinformatics analysis workflow2026
Modular large language model agents for multi-task computational materials science2026
Agentic AI in Healthcare and Medicine: A Seven-Dimensional Taxonomy for Empirical Evaluation of LLM-Based Agents2026
Autonomous in-silico inorganic materials discovery via multi-agent physics-aware scientific reasoning2026
AI agents in drug discovery: applications and case studies2026
An LLM-based multi-agent system to assist early-stage product design and evaluation2026