Theodore R. Sumers

Papers

Total Citations

H-Index

About

Theodore R. Sumers investigates how humans and machines can communicate more effectively, focusing on the intersection of robotics, artificial intelligence, and cognitive science. His primary research areas include language-guided robot learning, human-robot interaction, and abstraction learning—specifically, how robots can infer and leverage human preferences to generalize tasks beyond narrow demonstrations. Sumers’s most cited work, "Preference-Conditioned Language-Guided Abstraction" (2024), tackles a critical challenge in learning from demonstration: spurious correlations that limit generalization. He proposes that language can guide robots to construct task-relevant visual abstractions, enabling more robust, preference-aware learning. This contribution, already garnering 5 citations shortly after publication, highlights his ability to address foundational problems in robot learning. Sumers’s research is notable for bridging natural language processing and robotics, offering a path toward more intuitive human-robot teaching. His work has implications for assistive robotics, autonomous systems, and human-AI collaboration, positioning him as an emerging leader in making robots more adaptable and aligned with human intent.