Special Note: I’m currently hiring a Research Scientist or Postdoc to work in scientific feasibility assessment, with more info available here. I’m also looking for a student researcher (see here). Both are available immediately.

I am a broadly interdisciplinary artificial intelligence researcher specializing in natural language processing. I generally work in automated scientific reasoning (AI for science). I have a joint appointment between the University of Arizona College of Information Science and the Allen Institute for Artificial Intelligence (Ai2).

In terms of automated scientific discovery, some of my recent work includes automatically generating code-based experiments (e.g. CodeScientist), automatically generating scientific theories from literature (e.g. Theorizer), how to benchmark broad aspects of scientific discovery (e.g. AstaBench), automatically generating better and more useful hypotheses (e.g. HARPA), how we can perform automated scientific feasibility assessment (e.g. Matter-of-Fact and CodeDistiller), and large-scale knowledge resources for scientific discovery (e.g. the Scientific Contribution Graph). A recent talk (April 2026) on some of these projects is available on Youtube.

In terms of virtual environments, I explore how we can use text-based video games as a vehicle for studying scientific and common-sense reasoning, particularly by creating new virtual environments to encourage and measure these capacities for reasoning. These include projects like ScienceWorld, a high-fidelity simulation of 30 science exam experiments, TextWorldExpress, a text-game simulator capable of running up to one billion experiments per day, and DiscoveryWorld, a 2D (and text-based) game about making novel scientific discoveries on “Planet X”.

I’ve also worked extensively in methods of automated inference that produce explanations for their reasoning, particularly in the context of scientific knowledge and common-sense reasoning skills.  The Explanation Bank contains many such resources for explanation-centered inference, including WorldTree and EntailmentBank (book).

I uniquely have two distinct educational backgrounds, one in natural language processing, cognition, and computer science, the other in physics, electrical engineering, and sensing.  I maintain active outreach in grounding science education through sensing, largely in the form of open source hardware like the tricorder project, and projects like the open source computed tomography scanner. This work has been widely featured in over 50 international news media articles, including Reuters, Forbes, WIRED, MSNBC, and the Washington Post, as well as an invited talk at TEDxBrussels 2012.  In 2015, my open source science tricorder was honoured by being placed on permanent exhibit at the German Museum of Technology in Berlin.

Associate Professor, College of Information Science, University of Arizona
Cross-listed in Computer Science and Linguistcs.
office: harvill 437C
e-mail: pajansen@arizona.edu

Visiting Research Scientist, Allen Institute for Artificial Intelligence (Ai2)
e-mail: peterj@allenai.org