About
Joseph leads the CAMP Lab, where the goal is to build agents capable of lifelong learning.
A key piece of this is the ability to introspect, to understand their own knowledge and capabilities.
He explores this through interpretability, sample-efficient learning, theory of mind, and human–robot interaction.
Research interests
Lifelong learningInterpretabilityReinforcement learningTheory of mindHuman–robot interaction
Selected publications
Bayesian Social Deduction with Graph-Informed Language Models
S. Rahimirad, G. Gergerli, L. Romero, A. Qian, M. Olson, S. Stepputtis, J. Campbell
ACL 2026
Energy-Based Transfer for Reinforcement Learning
Z. Deng, J. Ghosh, F. Xie, K. Cheng, Y. Lu, K. Sycara, J. Campbell
CoLLAs 2026
Model-Agnostic Policy Explanations with Large Language Models
X. Zhang, Y. Guo, S. Chen, S. Stepputtis, M. Gombolay, K. Sycara, J. Campbell
COLM 2026