promptadmin/life-science-ai-prompts
1
0
Fork 0
Code Issues Pull Requests 48 Packages Projects Releases Wiki Activity
Files
main
life-science-ai-prompts/upstream/zhoujieli-Awesome-LLM-Agents-Scientific-Discovery
T
History

README.md

title, task, lineage_type, upstream_source, upstream_sha, imported_at, prompt_class, upstream_changes, author, validated
title task lineage_type upstream_source upstream_sha imported_at prompt_class upstream_changes author validated
Awesome AI Agents for Scientific Discovery [![Awesome](https://awesome.re/badge.svg)](https://awesome.re) import https://github.com/zhoujieli/Awesome-LLM-Agents-Scientific-Discovery/blob/bb3be5bc/README.md bb3be5bc 2026-08-08 prompt accepted upstream false

Awesome AI Agents for Scientific Discovery Awesome

AI Agents for Scientific Discovery

A curated list of papers about AI agents for scientific discovery and research automation.

Maintained by Jieli Zhou

If you use this paper list for your research, please cite it using:

@misc{zhou2024awesome,
  title={Awesome AI Agents for Scientific Discovery},
  author={Zhou, Jieli},
  year={2024},
  publisher={GitHub},
  journal={GitHub repository},
  howpublished={\url{https://github.com/zhoujieli/Awesome-LLM-Agents-Scientific-Discovery}}
}

Introduction

The convergence of large language models (LLMs) and autonomous agents has ushered in a new era in scientific discovery, fundamentally transforming how research is conducted across disciplines. This emerging paradigm, articulated in Kitano's seminal "Nobel Turing Challenge" (2021), envisions AI systems capable of making scientific discoveries worthy of Nobel Prize recognition. Recent advances in LLM-based agents have brought us closer to this vision, enabling increasingly sophisticated automation of scientific workflows and decision-making processes.

Evolution and Current Landscape

The field has evolved rapidly since early visions of AI-driven scientific discovery. While traditional AI systems focused on narrow tasks, modern LLM-based agents demonstrate remarkable capabilities in complex scientific reasoning, experimental design, and hypothesis generation. The breakthrough capabilities of models like GPT-4 have catalyzed this transition, enabling agents to engage in sophisticated scientific discourse, interpret complex data, and even design novel experiments.

Key Research Directions

Several major research themes have emerged in this space:

  1. Multi-Agent Architectures: Research has increasingly focused on collaborative multi-agent systems, where specialized agents work together to tackle complex scientific problems.

  2. Domain-Specific Applications: The healthcare sector has seen particularly rapid adoption, with agents being developed for clinical decision support, medical diagnosis, and healthcare administration.

  3. Scientific Process Automation: Agents are being developed to automate various aspects of the research pipeline, from literature review and hypothesis generation to experimental design and data analysis.

Impact and Future Directions

The emergence of AI agents in scientific discovery represents more than just technological advancement; it signals a fundamental shift in how science is conducted. These systems promise to:

  • Accelerate the pace of scientific discovery
  • Enable exploration of previously intractable research questions
  • Democratize access to scientific expertise
  • Foster more efficient use of research resources

Awesome LLM Agents for Scientific Discovery Awesome

AI Agents for Scientific Discovery

A curated list of papers about AI agents for scientific discovery and research automation.

Maintained by Jieli Zhou

If you use this paper list for your research, please cite it using:

@misc{zhou2024awesome,
  title={Awesome AI Agents for Scientific Discovery},
  author={Zhou, Jieli},
  year={2024},
  publisher={GitHub},
  journal={GitHub repository},
  howpublished={\url{https://github.com/zhoujieli/Awesome-LLM-Agents-Scientific-Discovery}}
}

Introduction

The convergence of large language models (LLMs) and autonomous agents has ushered in a new era in scientific discovery, fundamentally transforming how research is conducted across disciplines. This emerging paradigm, articulated in Kitano's seminal "Nobel Turing Challenge" (2021), envisions AI systems capable of making scientific discoveries worthy of Nobel Prize recognition. Recent advances in LLM-based agents have brought us closer to this vision, enabling increasingly sophisticated automation of scientific workflows and decision-making processes.

Evolution and Current Landscape

The field has evolved rapidly since early visions of AI-driven scientific discovery. While traditional AI systems focused on narrow tasks, modern LLM-based agents demonstrate remarkable capabilities in complex scientific reasoning, experimental design, and hypothesis generation. The breakthrough capabilities of models like GPT-4 have catalyzed this transition, enabling agents to engage in sophisticated scientific discourse, interpret complex data, and even design novel experiments.

Key Research Directions

Several major research themes have emerged in this space:

  1. Multi-Agent Architectures: Research has increasingly focused on collaborative multi-agent systems, where specialized agents work together to tackle complex scientific problems.

  2. Domain-Specific Applications: The healthcare sector has seen particularly rapid adoption, with agents being developed for clinical decision support, medical diagnosis, and healthcare administration.

  3. Scientific Process Automation: Agents are being developed to automate various aspects of the research pipeline, from literature review and hypothesis generation to experimental design and data analysis.

Impact and Future Directions

The emergence of AI agents in scientific discovery represents more than just technological advancement; it signals a fundamental shift in how science is conducted. These systems promise to:

  • Accelerate the pace of scientific discovery
  • Enable exploration of previously intractable research questions
  • Democratize access to scientific expertise
  • Foster more efficient use of research resources

Table of Contents

  1. Foundations & Vision
  2. Core Technologies
  3. Scientific Process Automation
  4. Domain Applications
  5. Infrastructure & Tools
  6. AI Agent Frameworks & Tools
  7. Evaluation & Benchmarking
  8. Surveys & Reviews

Foundations & Vision

Vision Papers

Core Technologies

Multi-Agent Systems & Architectures

Reasoning & Knowledge Systems

Scientific Process Automation

Research Planning & Literature Review

Experimental Design & Workflow

Domain Applications

Healthcare & Medicine

Clinical Decision Support & Diagnosis

Healthcare Systems & Management

Medical Education & Training

Medical Imaging & Pathology

Medical Research

Biology & Life Sciences

Genomics & Molecular Biology

Bioinformatics Tools & Platforms

Chemistry & Materials Science

Drug Discovery & Development

Molecular Modeling & Computation

Earth & Environmental Sciences

Evaluation & Benchmarking

General Benchmarks

Domain-Specific Benchmarks

Surveys & Reviews

Comprehensive Surveys

Domain-Specific Reviews

AI Agent Frameworks & Tools

  • Bride — Cognitive AI agent with Active Inference, HDC, and anomaly detection for hypothesis generation. [Rust, MIT] 2026

Contributing

Please feel free to send a pull request if you want to:

  • Add new papers
  • Fix errors
  • Update paper information

License

CC0

Reference in New Issue View Git Blame Copy Permalink