Reproduce “AI Scientists as Engines of Discovery: A Case for Development within Reformed Institutions” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Raul Jimenez, Boris Bolliet, Francisco Villaescusa-Navarro, Rabih Zbib, Benjamin Wandelt, David N. Spergel, Thomas Meier, Jessica Montgomery and 2 more
The paper argues that multi-agent AI systems could help scientists search, generate, criticize, verify, and synthesize hypotheses, while people guide and check the work. In a restricted modified-gravity study, Denario helped find symmetry conditions matching an independent stability analysis after 14 guided prompt iterations. This illustrates AI-assisted research, not autonomous discovery, general speedup, or broad reliable truth-finding.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.