Part 1: How close are language models to becoming autonomous and trustworthy scientists?
Presenter: Peter Jansen (Associate Professor at the University of Arizona, and Visiting Research Scientist at the Allen Institute for Artificial Intelligence)
Abstract: This talk examines the broader question of how close language models are to becoming autonomous, general-purpose scientists. We will consider this question through three recent projects, each addressing different aspects of the scientific discovery process. The first, DiscoveryWorld, studies whether AI scientists can complete the full cycle of scientific discovery in a virtual environment: generating hypotheses, designing and conducting experiments, analyzing results, and iterating toward novel discoveries -- all in a virtual game world that takes place on a hypothetical "Planet X". The second, CodeScientist, focuses on code-based discovery, investigating how well AI scientists can autonomously propose research hypotheses, write Python code to test them, and draft scientific papers when they identify potentially meaningful findings. The third, Theorizer, asks whether AI scientists can operate at a higher level of abstraction by inferring scientific theories from large collections of papers reporting experimental results. Taken together, these projects provide a lens on both the emerging strengths of current models and the substantial distance that still separates AI scientists from human domain experts.
Bio: Peter Jansen is an interdisciplinary AI researcher specializing in natural language processing, automated inference, and virtual world simulators, with a particular focus on automated scientific discovery. He holds a joint appointment as Associate Professor in the College of Information Science at the University of Arizona and Visiting Research Scientist at the Allen Institute for Artificial Intelligence (Ai2). His recent work on automated scientific discovery includes generating code-based experiments, synthesizing scientific theories from literature, benchmarking scientific reasoning, and assessing scientific feasibility, through projects such as CodeScientist, Theorizer, AstaBench, and Matter-of-Fact. He has also developed virtual environments for studying scientific reasoning, including ScienceWorld and DiscoveryWorld.
Part 2: Will Humans Make the Greatest Astronomy Discoveries of the Future?
Presenter: Ann Zabludoff (Professor of Astronomy, University of Arizona)
Abstract: I will review the broad range of work utilizing ML/AI for astronomy research at the University of Arizona and how our scientific questions push AI forward. I will discuss some of the limitations, including those encountered by our graduate student researchers and the critical need in our community for improved search and synthesis of the literature. I will discuss a possible path toward automated hypothesis generation through application of causal reasoning to structured and unstructured data extracted from scientific papers.
Bio: Ann Zabludoff has led studies across astronomy, astrophysics, and cosmology, including analyses of large observational datasets and theoretical simulations. She has worked to adapt astronomical instruments for new science. After obtaining S.B. degrees in Physics and in Mathematics from MIT, she received a Ph.D. in Astronomy from Harvard University. She was a Guggenheim Fellow, TEDx speaker, and the Caroline Herschel Distinguished Visitor at the Space Telescope Science Institute. She is a member of the UA College of Science Steering Committee for Data Science, Machine Learning, and AI and co-leads the Computation and Data Initiative of UA’s Theoretical Astrophysics Program. She is U.S. Participating Scientist on the Ultraviolet Transient Astronomy Satellite (ULTRASAT) science team. She has advised the National Science Foundation, NASA, the Department of Energy, and international research centers on science, policy, and prioritization.
Venue: Main Conference Room (MCR), NOIRLab, Tucson.
Date: Wednesday, April 29th, 2026.
11am CT-12pm CT (9am-10am MST)
Meeting ID: 847 4220 3545
Passcode: 634124