Friedrich Neubarth

Austrian Research Institute for Artificial Intelligence

Papers

Total Citations

H-Index

About

Friedrich Neubarth is a researcher at the forefront of human-robot interaction and grounded language learning, specializing in how robots acquire language through social and physical interaction with humans. His work centers on developing computational models that enable robots to learn word-object and word-action mappings from natural, cross-modal demonstrations, bridging the gap between human teaching behaviors and machine understanding. Neubarth’s most cited paper (11 citations) investigates transparency methods in a robot word-learning system, revealing how these methods influence naive users’ teaching strategies—a critical insight for designing intuitive, user-friendly robotic tutors. His 2018 demonstration on the Pepper robot showcased real-time grounded word learning, where the robot learned actions like “take” and “push” from a human tutor, highlighting practical applications in task-oriented scenarios. Additionally, his 2020 paper presents Bayesian and cross-situational models for incremental word learning, enabling robots to co-learn object-word mappings and referential intentions from minimal examples. With a growing citation impact, Neubarth’s contributions are shaping more adaptive, socially aware robots, making him a key figure in advancing human-robot collaboration and interactive AI.