From Automation to Autonomy: A Survey on Large Language Models in Scientific Discovery
What this work does
Classifies LLM roles as tools, analysts and scientists across stages of the scientific method, pairing methods with benchmark and domain examples.
Connection to discovery
A navigation framework for the awesome list and for separating individual research-stage competence from end-to-end autonomy.
Evidence
The inspected taxonomy covers hypothesis generation, experiment execution, analysis and refinement with references to original systems.
Scope and limitations
A survey label is not an empirical capability score. Each cited system still requires its own audit of training, evidence access and scientific validation.
THE RESEARCH CONVERSATION
Discussion & reactions
Share a question, an insight, or a replication experience. Sign in with GitHub to join the conversation.