Emergent autonomous scientific research capabilities of large language models

2023/04/11 by Daniil A. Boiko, Boiko, Daniil A., Robert MacKnight +3 · 5 voices · 13 citations
Computer Science · #Topic Modeling #cs.CL #physics.chem-ph

paper · pdf · doi:10.48550/arxiv.2304.05332

openalex publication_date 2023/04/11 · openalex created_date 2023/04/13 · openalex updated_date 2026/07/28

Abstract

Transformer-based large language models are rapidly advancing in the field of machine learning research, with applications spanning natural language, biology, chemistry, and computer programming. Extreme scaling and reinforcement learning from human feedback have significantly improved the quality of generated text, enabling these models to perform various tasks and reason about their choices. In this paper, we present an Intelligent Agent system that combines multiple large language models for autonomous design, planning, and execution of scientific experiments. We showcase the Agent's scientific research capabilities with three distinct examples, with the most complex being the successful performance of catalyzed cross-coupling reactions. Finally, we discuss the safety implications of such systems and propose measures to prevent their misuse.

Cited by

• Algorithmic Blindness in Large Language Models: A Calibration Study of Performance Prediction
• AInsteinBench: Benchmarking Coding Agents on Scientific Repositories
• AFRAgent : An Adaptive Feature Renormalization Based High Resolution Aware GUI agent
• Training-Free Active Learning Framework in Materials Science with Large Language Models
• DiscoVerse: Multi-Agent Pharmaceutical Co-Scientist for Traceable Drug Discovery and Reverse Translation
• Fetch.ai: An Architecture for Modern Multi-Agent Systems
• Chain-of-Trigger: An Agentic Backdoor that Paradoxically Enhances Agentic Robustness
• AutoLabs: Cognitive Multi-Agent Systems with Self-Correction for Autonomous Chemical Experimentation
• Can AI Follow In Einstein's Footsteps?
• Foam-Agent 2.0: An End-to-End Composable Multi-Agent Framework for Automating CFD Simulation in OpenFOAM
• Beyond Data Privacy: New Privacy Risks for Large Language Models
• Comp-X: On Defining an Interactive Learned Image Compression Paradigm With Expert-driven LLM Agent
• Experiences of Academics, Graduates, and Undergraduates in Using Generative AI in Research (Un)Ethically and (Ir)Responsibly: A Systematic Review of Qualitative Synthesis

Discussions

• Emergent autonomous scientific research capabilities of large language models [hn, 4 points, 0 comments]
• Emergent Autonomous Scientific Research Capabilities of LLMs (AI) [hn, 4 points, 0 comments]
• Emergent autonomous scientific research capabilities of large language models [hn, 3 points, 0 comments]
• Emergent autonomous scientific research capabilities of large language models [hn, 2 points, 0 comments]
• Emergent autonomous scientific research capabilities of large language models [hn, 1 points, 0 comments]

Related

• Autonomous chemical research with large language models
• Autonomous Agents for Scientific Discovery: Orchestrating Scientists, Language, Code, and Physics
• Advancing the Scientific Method with Large Language Models: From Hypothesis to Discovery
• A Review of Large Language Models and Autonomous Agents in Chemistry
• Large Language Model Agent: A Survey on Methodology, Applications and Challenges